ICAART 2013 Abstracts


Area 1 - Artificial Intelligence

Full Papers
Paper Nr: 25
Title:

Monitoring Agents in Complex Products - Enhancing a Discovery Robot with an Agent for Monitoring, Maintenance and Disaster Prevention

Authors:

Leo van Moergestel, Erik Puik, Daniël Telgen, Hendrik Folmer, Matthijs Grünbauer, Robbert Proost, Hielke Veringa and John-Jules Meyer

Abstract: Monitoring of computernetworks, complex technical systems like aeroplanes is common practice. In this paper we discuss the use of a monitoring agent in a product. The product itself could be any product with sufficient hardware capabilities, but the focus is on the product enhancement by adding an embedded agent. This socalled product agent can be a member of a multiagent system. In this way exchange of parts and subsystems is possible. The possibilities and advantages of this concept are discussed as well as a more elaborate example of the implementation in an experimental discovery robot.

Paper Nr: 34
Title:

Quantified Epistemic and Probabilistic ATL

Authors:

Henning Schnoor

Abstract: We introduce QAPI (quantified ATL with probabilism and incomplete information), which extends epistemic and probabilistic ATL with a flexible mechanism to reason about strategies in the object language, allowing very flexible treatment of the behavior of the “counter-coalition”. QAPI can express complex strategic properties such as equilibria. We show how related logics can be expressed in QAPI, provide bisimulation relations, and study the issues arising from the interplay between quantifiers and both epistemic and temporal operators.

Paper Nr: 39
Title:

A Graph-based Disambiguation Approach for Construction of an Expert Repository from Public Online Sources

Authors:

Anna Hristoskova, Elena Tsiporkova, Tom Tourwé, Simon Buelens, Mattias Putman and Filip De Turck

Abstract: The paper describes a dynamic framework for the construction and maintenance of an expert-finding repository through the continuous gathering and processing of online information. An initial set of online sources, relevant to the topic of interest, is identified to perform an initial collection of author profiles and publications. The extracted information is used as a seed to further enrich the expert profiles by considering other, potentially complementary, online data sources. The resulting expert repository is represented as a graph, where related author profiles are dynamically clustered together via a complex author disambiguation process leading to continuous merging and splitting of author nodes. Several rules are developed that assign weights to the links in the graph based on author similarities such as name, affiliation, e-mail, co-authors, and interests. Dynamic clustering of the authors depending on these weights results in the identification of unique experts for a specific domain. The developed disambiguation and author clustering algorithms are validated on several authors with varying name notations showing an improvement on the identification of unique profiles of 28% compared to the results from DBLP.

Paper Nr: 40
Title:

Algorithms for Acceptance in Argument Systems

Authors:

Samer Nofal, Paul Dunne and Katie Atkinson

Abstract: We introduce algorithms that decide arguments’ acceptance in Dung’s system of argumentation. Under \emph{preferred} semantics, there might be various extensions of acceptable arguments, and hence, the acceptance problem is concerned with deciding whether a given argument is in an extension or in all extensions. The new algorithms decide the acceptance without truly enumerating all extensions. This is of interest in situations where the acceptance problem is confined to a specific argument while the underlying argument system changes frequently such as in a dialog setting. We analyze our algorithms in contrast to existing algorithms. Consistent with experimental results, we argue that the new algorithms are more efficient with respect to running time.

Paper Nr: 45
Title:

Emergent Segmentation of Topological Active Nets by Means of Evolutionary Obtained Artificial Neural Networks

Authors:

Cristina V. Sierra, Jorge Novo, José Santos and Manuel G. Penedo

Abstract: We developed a novel segmentation method using deformable models. As deformable model we used Topological Active Nets, model which integrates features of region-based and boundary-based segmentation techniques. The deformation through time is defined by an Artificial Neural Network (ANN) that learns to move each node of the segmentation model based on its energy surrounding. The ANN is applied to each of the nodes and in different temporal steps until the final segmentation is obtained. The ANN training is obtained by simulated evolution, using differential evolution to automatically obtain the ANN that provides the emergent segmentation. The new proposal was tested in different artificial and real images, showing the capabilities of the methodology.

Paper Nr: 55
Title:

A Comparative Study of Different Image Features for Hand Gesture Machine Learning

Authors:

Paulo Trigueiros, Fernando Ribeiro and Luis Paulo Reis

Abstract: Vision-based hand gesture interfaces require fast and extremely robust hand detection, and gesture recognition. Hand gesture recognition for human computer interaction is an area of active research in computer vision and machine learning. The primary goal of gesture recognition research is to create a system, which can identify specific human gestures and use them to convey information or for device control. In this paper we present a comparative study of seven different algorithms for hand feature extraction, for static hand gesture classification, analysed with RapidMiner in order to find the best learner. We defined our own gesture vocabulary, with 10 gestures, and we have recorded videos from 20 persons performing the gestures for later processing. Our goal in the present study is to learn features that, isolated, respond better in various situations in human-computer interaction. Results show that the radial signature and the centroid distance are the features that when used separately obtain better results, being at the same time simple in terms of computational complexity.

Paper Nr: 87
Title:

Multi-criteria Evaluation of Class Binarization and Feature Selection in Tear Film Lipid Layer Classification

Authors:

Rebeca Méndez, Beatriz Remeseiro, Diego Peteiro-Barral and Manuel G. Penedo

Abstract: Dry eye is an increasingly popular syndrome in modern society which can be diagnosed through an automatic technique for tear film lipid layer classification. Previous studies related to this multi-class problem lack of analysis focus on class binarization techniques, feature selection and artificial neural networks. Also, all of them just use the accuracy of the machine learning algorithms as performance measure. This paper presents a methodology to evaluate different performance measures over these unexplored areas using the multiple criteria decision making method called TOPSIS. The results obtained demonstrate the effectiveness of the methodology proposed in this research.

Paper Nr: 89
Title:

Simulating Human Activities to Investigate Household Energy Consumption

Authors:

Édouard Amouroux, Thomas Huraux, François Sempé, Nicolas Sabouret and Yvon Haradji

Abstract: This paper presents a multi-agent model and simulator for the investigation of inhabitants’ behaviour in relation with energy consumption in a household, at a fine-grain level. We present the meta-model that allows non-computer specialists to define the household activities, and the SMACH simulator that helps specialists analyse behaviour and manage and control energy consumption. We illustrate on concrete example scenarios how the system’s adaptation mechanisms allow us to outline behavioural patterns.

Paper Nr: 96
Title:

Enhancing Attentive Task Search with Information Gain Trees and Failure Detection Strategies

Authors:

Kristin Stamm and Andreas Dengel

Abstract: Enterprises today are challenged by managing requests arriving through all communication channels. To support service employees in better and faster understanding incoming documents, we developed the approach of process-driven document analysis (DA). We introduced the structure Attentive Task (AT) to formalize information expectations toward an incoming document. To map the documents to the corresponding AT, we previously developed a novel search approach that uses DA results as evidences for prioritizing all AT. With this approach, we consider numerous task instances including their context instead of a few process classes. The application of AT search in enterprises raises two challenges: (1) Complex domains require a structured selection of well performing evidence types, (2) a failure detection method is needed for handling a substantial part of incoming documents that cannot be related to any AT. Here, we apply methods from machine learning to meet these requirements. We learn and apply information gain trees for structuring and optimizing evidence selection. We propose five strategies for detecting documents without ATs. We evaluate the suggested methods with two processes of a financial institution.

Paper Nr: 118
Title:

Constraint-handling for Optimization with Support Vector Surrogate Models - A Novel Decoder Approach

Authors:

Jörg Bremer and Michael Sonnenschein

Abstract: A new application for support vector machines is their use for meta-modeling feasible regions in constrained optimization problems. We here describe a solution for the still unsolved problem of a standardized integration of such models into (evolutionary) optimization algorithms with the help of a new decoder based approach. This goal is achieved by constructing a mapping function that maps the whole unconstrained domain of a given problem to the region of feasible solutions with the help of the the support vector model. The applicability to real world problems is demonstrated using the load balancing problem from the smart grid domain.

Paper Nr: 127
Title:

Dynamic Selection of Learning Situations in Virtual Environment

Authors:

Kevin Carpentier, Domitile Lourdeaux and Indira Mouttapa-Thouvenin

Abstract: In a lot of industrial contexts, workers may encounter novel situations which have never occured in their training. Yet, such situations must be handeld successfully to prevent high-cost consequences. Such consequences might be human casualties (in high-risk domains), material damages (in manufacturing domains) or productivity loss (in high performance industry). To address this lack in their training, virtual environments for training should provide a large spectrum of learning situations. The difficulty lies in generating these situations dynamically according to the learners profile while they have a total freedom of interaction in the virtual environment. To address this issue, we propose to generate activities by operationnalising the Zone of Proximal Development in a multidimensional space. The filling of this space will be updated dynamically based on user interactions.

Paper Nr: 133
Title:

Compressing Multi-document Summaries through Sentence Simplification

Authors:

Sara Botelho Silveira and Antonio Branco

Abstract: Multi-document summarization aims at creating a single summary based on the information conveyed by a collection of texts. After the candidate sentences have been identified and ordered, it is time to select which will be included in the summary. In this paper, we propose an approach that uses sentence simplification, both lexical and syntactic, to help improve the compression step in the summarization process. Simplification is performed by removing specific sentential constructions conveying information that can be considered to be less relevant to the general message of the summary. Thus, the rationale is that sentence simplification not only removes expendable information, but also makes room for further relevant data in a summary.

Paper Nr: 148
Title:

Planning of Diverse Trajectories

Authors:

Jan Tožička, David Šišlákand and Michal Pěchouček

Abstract: Unmanned aerial vehicles (UAVs) are more and more often used to solve different tasks in both the private and the public sector. Some of these tasks can often be performed completely autonomously while others are still dependent on remote pilots. They control an UAV using a command display where they can control it manually using joysticks or give it a simple task. The command display allow for the planning of the UAV trajectory through waypoints while avoiding no-fly zones. Nevertheless, the operator can be aware of other preferences or soft restrictions for which it’s not feasible to be inserted into the system especially during time critical tasks. We propose to provide the operator with several different alternative trajectories, so he can choose the best one for the current situation. In this contribution we propose several metrics to measure the diversity of the trajectories. Then we explore several algorithms for the alternative trajectories creation. Experimental results in two grid domains show how the proposed algorithms perform.

Paper Nr: 149
Title:

Rule-based Behavioral Reasoning on Semantic Business Processes

Authors:

Fabrizio Smith and Maurizio Proietti

Abstract: We propose a rule-based framework for representing and reasoning about business processes from both the procedural and ontological point of views. To this end we define a rule-based procedural semantics for a relevant fragment of BPMN, a very popular graphical notation for specifying business processes. Our semantics defines a state transition system by following an approach similar to the Fluent Calculus, and allows us to specify state change in terms of preconditions and effects of the enactment of activities. Then we show how the procedural process knowledge can be seamlessly integrated with the domain knowledge specified by using the OWL-RL rule-based ontology language. Our framework provides a wide range of reasoning services by using standard logic programming inference engines. In particular, we can perform very sophisticated reasoning tasks by combining both procedural and domain dependent knowledge. A preliminary implementation shows that our approach is effective in practice.

Paper Nr: 157
Title:

An Algorithm for Checking the Dynamic Controllability of a Conditional Simple Temporal Network with Uncertainty

Authors:

Carlo Combi, Luke Hunsberger and Roberto Posenato

Abstract: A Simple Temporal Network with Uncertainty (STNU) is a framework for representing and reasoning about temporal problems involving actions whose durations are bounded but uncontrollable. A dynamically controllable STNU is one for which there exists a strategy for executing its time-points that guarantees that all of the temporal constraints in the network will be satisfied no matter how the uncontrollable durations turn out. A Conditional Simple Temporal Network with Uncertainty (CSTNU) augments an STNU to include observation nodes, where the execution of each observation node provides, in real time, the truth value of an associated proposition. Recent work has generalized the notion of dynamic controllability to cover CSTNUs. This paper presents an algorithm—called a DC-checking algorithm—for determining whether arbitrary CSTNUs are dynamically controllable. The algorithm, which is proven to be sound, is the first such algorithm to be presented in the literature. The algorithm extends edge-generation/constraint-propagation rules from an existing STNU algorithm to accommodate propositional labels, while adding new rules required to deal with the observation nodes. The paper also discusses implementation issues associated with the management of propositional labels.

Paper Nr: 190
Title:

Magic Loops in Simple Temporal Networks with Uncertainty - Exploiting Structure to Speed Up Dynamic Controllability Checking

Authors:

Luke Hunsberger

Abstract: A Simple Temporal Network with Uncertainty (STNU) is a structure for representing and reasoning about temporal constraints and uncontrollable-but-bounded temporal intervals called contingent links. An STNU is dynamically controllable (DC) if there exists a strategy for executing its time-points that guarantees that all of the constraints will be satisfied no matter how the durations of the contingent links turn out. The fastest algorithm in the literature for checking the dynamic controllability of arbitrary STNUs is based on an analysis of the graphical structure of STNUs. This paper (1) presents a new method for analyzing the graphical structure of STNUs, (2) determines an upper bound on the complexity of certain structures---the indivisible semi-reducible negative loops; (3) presents an algorithm for generating loops---the magic loops---whose complexity attains this upper bound; and (4) shows how the upper bound can be exploited to speed up the process of DC-checking for certain networks. Theoretically, the paper deepens our understanding of the structure of STNU graphs. Practically, it demonstrates new ways of exploiting graphical structure to speed up DC checking.

Paper Nr: 193
Title:

Evolving Urbanisation Policies - Using a Statistical Model to Accelerate Optimisation over Agent-based Simulations

Authors:

Marta Vallejo, David W. Corne and Verena Rieser

Abstract: Agent-based systems are commonly used in the geographical land use sciences to model processes such as urban growth. In some cases, agents represent civic decision-makers, iteratively making decisions about the sale, purchase and development of patches of land. Based on simple assumptions, such systems are able broadly to model growth scenarios with plausible properties and patterns that can support decision-makers. However, the computational time complexity of simulations limits the use of such systems. Attractive possibilities, such as the optimisation of urban growth policies, tend to be unexplored since the time required to run many thousands of simulations is unacceptable. In this paper we address this situation by exploring an approach that makes use of a statistical model of the agent-based system’s behaviour to inform a rapid approximation of the fitness function. This requires a limited number of prior simulations, and then allows the use of an evolutionary algorithm to optimise urban growth policies, where the quality of a policy is evaluated within a highly uncertain environment. The approach is tested on a typical urban growth simulation, in which the overall goal is to find policies that maximise the ’satisfaction’ of the residents. We find that the model-driven approximation of the simulation is effective at leading the evolutionary algorithm towards policies that yield vastly better satisfaction levels than unoptimised policies.

Paper Nr: 197
Title:

Level of Detail based AI Adaptation for Agents in Video Games

Authors:

Ghulam Mahdi, Yannick Francillette, Gouaich Abdelkader, Fabien Michel and Nadia Hocine

Abstract: This paper suggests multi-agent systems (MASs) for implementing game artificial intelligence (AI) for video games. One of main hindrances against using MASs technology in video games has been the real-time constraints for frame rendering. In order to deal with the real-time constraints, we introduce an adaptation-oriented approach for maintaining frame rate in acceptable ranges. The adaptation approach is inspired from the level of detail (LoD) technique in 3D graphics. We introduce agent organizations for defining different roles of agents in game AI. The computational requirements of agent roles have been prioritized according to their functional roles in a game. In this way, adapting computational requirements of game AI works as a means for maintaining frame rate in acceptable ranges. The proposed approach has been evaluated through a pilot experiment by using a proof of concept game. The pilot experiment shows that LoD based adaptation allows maintaining frame rate in acceptable ranges and therefore enhancing the quality of service.

Short Papers
Paper Nr: 17
Title:

Modal Semirings with Operators for Knowledge Representation

Authors:

Kim Solin

Abstract: Modal semirings are combined with modal algebra (Boolean algebra with operators) to form modal semirings with operators. In turn, these are extended with a revision operator and used for knowledge representation.

Paper Nr: 30
Title:

Reinforcement Learning for Multi-purpose Schedules

Authors:

Kristof Van Moffaert, Yann-Michaël De Hauwere, Peter Vrancx and Ann Nowé

Abstract: In this paper, we present a learning technique for determining schedules for general devices that focus on a combination of two objectives. These objectives are user-convenience and gains in energy savings. The proposed learning algorithm is based on Fitted-Q Iteration (FQI) and analyzes the usage and the users of a particular device to decide upon the appropriate profile of start-up and shutdown times of that equipment. The algorithm is experimentally evaluated on real-life data to discover that close-to-optimal control policies can be learned on a short timespan of a only few iterations. Our results show that the algorithm is capable of proposing intelligent schedules depending on which objective the user placed more or less emphasis on.

Paper Nr: 46
Title:

From Study of Human-human Dialogues to Reasoning Model - Conversational Agent in Argumentation Dialogue

Authors:

Mare Koit and Haldur Õim

Abstract: We study human-human dialogues where one of the participants tries to influence the reasoning process of the dialogue partner in order to force the partner to make a decision to perform an action. Our further aim is to implement a dialogue system which would interact with a user in natural language. A model of the motivational sphere of a reasoning subject will be presented as a vector which consists of evaluations of different aspects of the action under consideration. Three reasoning procedures will be introduced, each of which is triggered by a so-called input factor. We examine the communicative strategies and communicative tactics that dialogue participants use to achieve their communicative goals. The models are implemented as a computer program.

Paper Nr: 51
Title:

The State of the Art in the Development of a Versatile Argumentation System based on the Logic of Multiple-valued Argumentation

Authors:

Satoru Tannai, Shogo Ohta, Takeshi Hagiwara, Hajime Sawamura and Jacques Riche

Abstract: This paper reports on the state of the art in the development of a versatile argumentation system in which various auxiliary features for argumentation are incorporated. Such an argumentation system was built on our underlying argumentation system, LMA: the Logic of Multiple-valued Argumentation, aiming at promoting our understanding of argumentation processes. We in particular present some very new and unique aspects of computational argumentation: the syncretic argumentation, the argument mining and frequent sub-argument discovery, the argument metamorphosis from symbol to animation via natural language, Eastern argumentation based on Indian logic, and the argumentation based on symbolism and connectionism. Such a hybridization of various features would broaden the scope of the applications of computational argumentation in various ways.

Paper Nr: 52
Title:

Empirical Evaluation of Shortest Path Gaussian Kernels over State Action Graphs

Authors:

Saba Q. Yahyaa and Bernard Manderick

Abstract: We define shortest path Gaussian kernels basis functions over state graphs and state-action graphs. We empirically demonstrate that these new basis functions used in linear parametric function approximation outperform basis functions defined on the state space, the state graph and the state-action graph.

Paper Nr: 68
Title:

Dynamic Scenario Adaptation Balancing Control, Coherence and Emergence

Authors:

Camille Barot, Domitile Lourdeaux and Dominique Lenne

Abstract: As the industrial world grows more complex, virtual environments have proven to be interesting tools to train workers to procedures and work situations. To ensure learning and motivation from the trainees, a pedagogical control of these environments is needed. However, existing systems either do not provide control over running simulations, limit user agency, need the authors to specify a priori all possible scenarios, or allow incoherent behaviours from the simulated technical system or the virtual characters. We propose in this paper a model for a dynamic and indirect control of the events of a virtual environment. Our model aims to ensure the control, coherence, and emergence of situations, in virtual environments designed for training in highly complex work situations.

Paper Nr: 69
Title:

When you Talk about “Information Processing” What Actually Do you have in Mind?

Authors:

Emanuel Diamant

Abstract: “Information Processing” is a recently launched buzzword whose meaning is vague and obscure even for the majority of its users. The reason for this is the lack of a suitable definition for the term “information”. In my attempt to amend this bizarre situation, I have realized that, following the insights of Kolmogorov’s Complexity theory, information can be defined as a description of structures observable in a given data set. Two types of structures could be easily distinguished in every data set – in this regard, two types of information (information descriptions) should be designated: physical information and semantic information. Kolmogorov’s theory also posits that the information descriptions should be provided as a linguistic text structure. This inevitably leads us to an assertion that information processing has to be seen as a kind of text processing. The idea is not new – inspired by the observation that human information processing is deeply rooted in natural language handling customs, Lotfi Zadeh and his followers have introduced the so-called “Computing With Words” paradigm. Despite of promotional efforts, the idea is not taking off yet. The reason – a lack of a coherent understanding of what should be called “information”, and, as a result, misleading research roadmaps and objectives. I hope my humble attempt to clarify these issues would be helpful in avoiding common traps and pitfalls.

Paper Nr: 91
Title:

Cargo Transportation Models Analysis using Multi-Agent Adaptive Real-Time Truck Scheduling System

Authors:

Oleg Granichin, Petr Skobelev, Alexander Lada, Igor Mayorov and Alexander Tsarev

Abstract: The use of multi-agent platform for real-time adaptive scheduling of trucks is considered. The schedule in such system is formed dynamically by balancing the interests of orders and resource agents. The system doesn’t stop or restart to rebuild the plan of mobile resources in response to upcoming events but finds out conflicts and adaptively re-schedule demand-resource links in plans when required. Different organizational models of cargo transportation for truck companies having own fleet are analyzed based on simulation of statistically representative flows of orders. Models include the rigid ones, where trucks return back to their garage after each trip, and more flexible, where trucks wait for new orders at the unloading positions, where trucks can be late but pay a penalty for this, and finally where orders can be adaptively rescheduled ’on the fly‘ in real-time and the schedule of each truck can change individually during orders execution. Results of simulations of trucks profit depending on time period are presented for each model. These results show measurable benefits of using the multi-agent systems with real-time decision making - up to 40-60% comparing with rigid models. The profit dependencies on the number of trucks are also built and analyzed. The results show that using adaptive scheduling in real time it is possible to execute the same number of orders with less trucks (up to 20%).

Paper Nr: 108
Title:

Face Recognition under Real-world Conditions

Authors:

Ladislav Lenc and Pavel Král

Abstract: This paper deals with Automatic Face Recognition (AFR). The main contribution of this work consists in the evaluation of our two previously proposed AFR methods in real conditions. At first, we compare and evaluate the recognition accuracy of two AFR methods on well-controlled face database. Then we compare these results with the recognition accuracy on a real-world database of comparable size. For such comparison, we use a sub-set of the newly created Czech News Agency (ˇCTK) database. This database is created from the real photos acquired by the ˇCTK and the creation of this corpus represents the second contribution of this work. The experiments show the significant differences in the results on the controlled and real-world data. 100% accuracy is achieved on the ORL database while only 72.7% is the best score for the ˇCTK database. Further experiments show, how the recognition rate is influenced by the number of training images for each person and by the size of the database. We also demonstrate, that the recognition rate decreases significantly with larger database. We propose a confidence measure technique as a solution to identify and to filter-out the incorrectly recognized faces. We further show that confidence measure is very beneficial for AFR under real conditions.

Paper Nr: 123
Title:

Adaptive Agents for Cyber-Physical Systems

Authors:

Ichiro Satoh

Abstract: This paper proposes a bio-inspired approach to adapting software components in CPSs. It introduces the notions of differentiation and dedifferentiation in cellular slime molds. When a software component delegates a function to another component coordinating with it, if the former has the function, this function becomes lessdeveloped and the latter’s function becomes well-developed like that in cellular differentiation. The approach enables software components on CPSs to be naturally adapted to changes in the cyber and physical world in a self-organizing manner. It is constructed as a middleware system to execute general purpose applications on a CPS. We present several evaluations of the approach in CPSs.

Paper Nr: 134
Title:

Knowledge Gradient Exploration in Online Least Squares Policy Iteration

Authors:

Saba Q. Yahyaa and Bernard Manderick

Abstract: We compare empirically the knowledge gradient exploration policy with the e-greedy one in online leastsquares policy iteration on a testbed of 2 infinite horizon Markov decision problems. It is shown that the knowledge gradient, although it does not have parameters to be tuned, performs as well as a well-tuned e-greedy exploration policy.

Paper Nr: 142
Title:

Data Mining for Real-Time Intelligent Decision Support System in Intensive Care Medicine

Authors:

Filipe Portela, Manuel Filipe Santos, Álvaro Silva, José Machado, António Abelha and Fernando Rua

Abstract: The introduction of Intelligent Decision Support Systems (IDSS) in critical areas like Intensive Medicine is a complex and difficult process. The professionals of Intensive Care Units (ICU) haven’t much time to register data because the direct care to the patients is always mandatory. In order to help doctors in the decision making process, the INTCare system has been deployed in the ICU of Centro Hospitalar of Porto, Portugal. INTCare is an IDSS that makes use of data mining models to predict the outcome and the organ failure probability for the ICU patients. This paper introduces the work carried out in order to automate the processes of data acquisition and data mining. The main goal of this work is to reduce significantly the manual efforts of the staff in the ICU. All the processes are autonomous and are executed in real-time. In particular, Decision Trees, Support Vector Machines and Naïve Bayes were used with online data to continuously adapt the predictive models. The data engineering process and achieved results, in terms of the performance attained, will be presented.

Paper Nr: 155
Title:

Efficient Online Feature Selection based on ℓ1-Regularized Logistic Regression

Authors:

Kengo Ooi and Takashi Ninomiya

Abstract: Finding features for classifiers is one of the most important concerns in various fields, such as information retrieval, speech recognition, bio-informatics and natural language processing, for improving classifier prediction performance. Online grafting is one solution for finding useful features from an extremely large feature set. Given a sequence of features, online grafting selects or discards each feature in the sequence of features one at a time. Online grafting is preferable in that it incrementally selects features, and it is defined as an optimization problem based on ℓ1-regularized logistic regression. However, its learning is inefficient due to frequent parameter optimization. We propose two improved methods, in terms of efficiency, for online grafting that approximate original online grafting by testing multiple features simultaneously. The experiments have shown that our methods significantly improved efficiency of online grafting. Though our methods are approximation techniques, deterioration of prediction performance was negligibly small.

Paper Nr: 156
Title:

Plan Synthesis for Probabilistic Activity Recognition

Authors:

Frank Krüger, Kristina Yordanova, Albert Hein and Thomas Kirste

Abstract: We analyze the applicability of model-based approaches to the task of inferring activities in smart environments. We introduce a symbolic approach to representing human behavior that enables the use of prior knowledge on the causality of human action and outline its probabilistic semantics. Based on an experimental analysis of a real world scenario from the smart meeting room domain, we show that such a symbolic approach allows to build reusable behavior models that compete with data-driven models at the performance level and that are able to track human behavior across a wide range of scenarios.

Paper Nr: 165
Title:

A Fuzzy Dynamic Belief Logic

Authors:

Xiaoxin Jing and Xudong Luo

Abstract: This paper introduces a new logic approach to reason about the dynamic belief revision by extending wellknown Aucher’s dynamic belief revision approach to fuzzy environments. In our system, propositions take a valuation in linguistic truth term set, and a belief revision is also in a qualitative way. Moreover, we reveal some properties of our system in epistemic style and do a comparison between our fuzzy belief system with famous AGM postulats.

Paper Nr: 172
Title:

F T E: A Fuzzy Timed Action Language

Authors:

Youzhi Zhang, Xudong Luo and Yuping Shen

Abstract: This paper proposes a fuzzy approach for reasoning about action and change in timed domains. In our method, actions and world states are modeled as fuzzy sets over time axis. Thus, their temporal relations and time constraints can be model as fuzzy rules. So, our method handles well the issue that action happens at an approximate time and then the states change also at an approximate time, which has not been solved well in existing work. Finally, our method is used to solve the classic problem of rail road crossing control in a fuzzy environment. The theoretic and simulation analysis shows that the controller using our method works well

Paper Nr: 177
Title:

An Efficient Translation Scheme for Representing Nurse Rostering Problems as Satisfiability Problems

Authors:

Stefaan Haspeslagh, Tommy Messelis, Greet Vanden Berghe and Patrick De Causmaecker

Abstract: In this paper we present efficient translation schemes for converting nurse rostering problem instances into satisfiability problems (SAT). We define eight generic constraints types allowing the representation of a large number of nurse rostering constraints commonly found in literature. For each of the generic constraint types, we present efficient translation schemes to SAT. Special attention is paid to the representation of counting constraints. We developed a two way translation scheme for counting constraints using O(nlogn) variables and O(n2) clauses. We translated the instances of the First international nurse rostering competition 2010 to SAT and proved the infeasibility of the instances. The SAT translation was used for a hardness study of nurse rostering problem instances based on SAT features.

Paper Nr: 191
Title:

Towards Epistemic Planning Agents

Authors:

Manfred Eppe and Frank Dylla

Abstract: We propose an approach for single-agent epistemic planning in domains with incomplete knowledge. We argue that on the one hand the integration of epistemic reasoning into planning is useful because it makes the use of sensors more flexible. On the other hand, defining an epistemic problem description is an error prone task as the epistemic effects of actions are more complex than their usual physical effects. We apply the axioms of the Discrete Event Calculus Knowledge Theory (DECKT) as rules to compile simple non-epistemic planning problem descriptions into complex epistemic descriptions. We show how the resulting planning problems are solved by our implemented prototype which is based on Answer Set Programming (ASP).

Paper Nr: 199
Title:

SPIG - Security by Privileges, Inspection and Guaranty for Ambient Agents Group

Authors:

Nardjes Bouchemal and Ramdane Maamri

Abstract: In a future vision of Ambient Intelligence – or AmI – our surrounding environment will integrate a pervasive and interconnected network of devices equipped with sensors, actuators and ambient agents. However, it is a big security issue when building this kind of complex systems, and it is not enough that ambient agent platform provides a set of standard security mechanisms such as sandboxing, encryption and digital signatures. Furthermore, we must take into account ambient agent limitations and we mustn’t endow it with complex cryptography concepts or historic data. This is why our approach, proposed in this paper, is based on cooperation between group members and behavior inspection of new integrated agents, in addition of some cryptography concepts. The proposed protocol is based on cooperation, collective decision, guarantee and attribution of privileges according to the trust degree of agents.

Paper Nr: 203
Title:

Enhanced Shortest Path Computation for Multiagent-based Intermodal Transport Planning in Dynamic Environments

Authors:

Christoph Greulich, Stefan Edelkamp, Max Gath, Tobias Warden, Malte Humann, Otthein Herzog and T. G. Sitharam

Abstract: This paper addresses improved urban mobility using multiagent simulation. We provide a description of the agent model and the routing infrastructure as a step towards a rich model of the interactions that happen in intermodal transport planning tasks. The multiagent model is generic in the sense that different public and individual transport agents and transportation agencies can be added and parameterized on-the-fly. It integrates planning with execution. We show that a sequence of calls to Dijkstra’s single-source shortest-paths algorithm is crucial for planning and provide an efficient memory-less implementation with radix heaps in order to make this application feasible with respect to scalability. As a case study, we implement a scenario for Bangalore (India), starting on a higher level of abstraction and drilling down to a running program.

Paper Nr: 213
Title:

Constrained Minimum-Variance PID Control using Hybrid Nelder-Mead Simplex and Swarm Intelligence

Authors:

N. Pillay and P. Govender

Abstract: The paper proposes the use of an efficient hybrid optimization routine by combining Nelder-Mead simplex with Particle Swarm algorithm (NMPSO) to synthesize a proportional-integral-derivative (PID) type controller. The conceived controller is capable of providing the best possible performance for regulating stochastic disturbances under closed loop conditions. A global optimal solution is found by exploiting the process output variance expression in terms of its closed loop impulse response coefficients. The results of which are used to define an achievable lower bound of the PID performance in terms of the output variance of the closed loop system. Several simulation examples drawn from literature are used to demonstrate the efficacy of the proposed methodology.

Paper Nr: 221
Title:

Towards a Sustainable Smart e-Marketplace - A Stable, Efficient and Responsive Smart Exchange with Strategic Conduct

Authors:

Wafa Ghonaim, Hamada Ghenniwa and Weiming Shen

Abstract: The landscapes of e-marketplaces are changing profoundly, evident in the phenomenal growth and potential of online services, consumers, and enabling mobile technologies. However, it is unleashing grave concerns about sustainability due to the fierce competitions, fuzzy dynamics and rapidly shifting powers. While it is attributed to the game-theoretic economics and computation complexities of the decentralized combinatorial allocation problem, this work establishes, denying e-traders expressing fair strategic choice is unfounded of adverse strategic risk. In fact, free market dynamics realize impact of smart learning on strategic conduct. The fact strategic rules enable faster consumer-to-market bidding lifecycle is another compelling factor. Hence, the work introduces the novel rule-based bidding language and GSPM double auction for the smart exchange that facilitates expressions of strategic rules, while uniquely exploits forward and reverse GSP auctions for efficient, tractable, stable, and budget balanced e-marketplace. The e-marketplace deliberates on rules for effective preference elicitation, while bringing self-prosperity in socially efficient ecosystem.

Paper Nr: 228
Title:

A Top-down Approach to Combining Logics

Authors:

Christoph Benzmüller

Abstract: The mechanization and automation of combination of logics, expressive ontologies and notions of context are prominent current challenge problems. I propose to approach these challenge topics from the perspective of classical higher-order logic. From this perspective these topics are closely related and a common, uniform solution appears in reach.

Paper Nr: 233
Title:

User Profiling of People with Disabilities - A Proposal to Pervasively Assess Quality of Life

Authors:

Eloisa Vargiu, Luigi Ceccaroni, Laia Subirats, Suzanne Martin and Felip Miralles

Abstract: This paper presents and discusses an ongoing work aimed at defining the profile of people with disabilities, i.e., automatically assessing their quality of life, through a sensor-based telemonitoring system. To illustrate how the approach works, a case study is presented and discussed.

Paper Nr: 234
Title:

Two Sides of a Coin - Translate while Classify Multilanguage Annotations with Domain Ontology-driven Word Sense Disambiguation

Authors:

Massimiliano Gioseffi and Angela Locoro

Abstract: In this paper we present an approach for the translation and classification of short texts in one step. Our work lays in the tradition of Domain-Driven Word Sense Disambiguation, though a major emphasis is given to domain ontologies as the right tool for sense-tagging and topic detection of short texts which, by their nature, are known to be reluctant to statistical treatment. We claim that in a scenario where users can annotate knowledge items using different languages, domain ontologies can prove very suitable for driving the word disambiguation and topic classification tasks. In this way, two tasks are gainfully collapsed in a single one. Although this study is still in its infancy, in what follows we are able to articulate motivations, design, workflow analysis, and concrete evolutions envisioned for our tool.

Paper Nr: 237
Title:

Ontologies for Authoring, or Authoring Ontologies?

Authors:

Diana Arellano, Javier Varona and Volker Helzle

Abstract: In the last years the use of ontologies has broaden to areas that until some time ago were unthinkable, like storytelling or context/content representation. The main problem with the use of ontologies is that the user responsible of authoring the story needs to input every single element that is required for the story to make sense. Depending on the case, this might be a tedious task. However, once it is done, different stories can be developed by reusing the already defined concepts. The objective of the paper is to provide examples of applications where the use of ontologies conveyed “authoring effort” with satisfactory results. We also state our opinion of why is it better to use ontologies for such tasks, explain our own experience with an use case and propose ideas of what could be enhanced, or taken from other areas, to improve the authoring process.

Paper Nr: 240
Title:

Group Recommender Systems - Some Experimental Results

Authors:

Vineet Padmanabhan, Prabhu Kiran and Abdul Sattar

Abstract: Recommender Systems (RS) are software applications which aim to support users in their decision making while interacting with large information spaces. Most recommender systems are designed for recommending items to individuals. In this paper we provide experimental results related to developing a content-based group recommender system. To this end we make two important contributions. (1) Implementation of a group recommender system based on as proposed recently in vineet et.al. using MovieLens dataset which is a relatively huge data-set (100,000 ratings from 943 users on 1682 movies) as compared to the data-set size of 150 used in vineet et.al. (2) We use seven variants of decision-tree measures and built an empirical comparison table to check for precision rate in group recommendations based on different social-choice theory strategies.

Paper Nr: 244
Title:

Agents and Analytics - A Framework for Educational Data Mining with Games based Learning

Authors:

Harri Ketamo

Abstract: This paper focuses on data mining and analysis framework behind Eedu elements mathematics game. The background of the game is in learning-by-doing, learning-by-teaching and to some extent learning-by-programming. The data modelling behind the game is based on semantic networks. When all the skills and knowledge is modelled as semantic network, all the data mining can be done in terms of network analysis. According to our studies, this approach enables very detailed and valid learning analytics. The novelty value of the study is in games based approach on learning and data mining.

Paper Nr: 249
Title:

Can Fuzzy Decision Support Link Serial Serious Crime?

Authors:

Don Casey and Phillip Burrell

Abstract: The problem addressed is one of great practical significance in the investigation of stranger rape. The linkage of these crimes at an early stage is of the greatest importance in a successful prosecution and also in the prevention of further crimes that may be even more serious. One of the most important considerations when investigating a serious sexual offence is to find if it can be linked to other offences; if this can be done then there is a considerable dividend in terms additional evidence and new lines of enquiry. In spite of a great deal of research into this area and the expenditure of considerable resources by law enforcement agencies across the world there is no computer-based decision support system that assist crime analysts in this important task. A number of different crime typologies have been presented but their utility in decision support is unproven. It is the authors’ contention that difficulties arise from the inadequacy of the adoption of the classical or ‘crisp set’ paradigm. Complex events like crimes cannot be described satisfactorily in this way and it proposed that fuzzy set theory offers a powerful framework within which crime can be portrayed in a sensitive manner and that this can integrate psychological knowledge in order to enhance crime linkage.

Paper Nr: 252
Title:

A Long Term Proposal to Simulate Consciousness in Artificial Life

Authors:

Joseph D. Horton, Michael Francis and Eckart Sußenburger

Abstract: Computers will soon be powerful enough to simulate consciousness. The artificial life community should start to try to understand how consciousness could be simulated. The proposal is to build an artificial life system in which consciousness might be able to evolve. The idea is to develop internet-wide artificial universe in which the agents can evolve. Users play games by defining agents that form communities. The communities have to perform tasks, or compete, or whatever the specific game demands. The demands should be such that agents that are more aware of their universe are more likely to succeed. The agents reproduce and evolve within their user’s machine, but can also sometimes transfer to other machine across the internet. Users will be able to choose the capabilities of their agents from a fixed list, but may also write their own powers for their agents

Paper Nr: 260
Title:

Intercultural-role Plays for e-Learning using Emotive Agents

Authors:

Cat Kutay, Samuel Mascarenhas, Ana Paiva and Rui Prada

Abstract: This paper presents joint work between an Australian team developing role-based games for experiential learning of Aboriginal culture, and a Portuguese research department developing interactive modules to create believable agent reactions in virtual environments. The game incorporates recorded stories in an online system to teach culture. Teaching scenarios group the narratives along a learning path, but their presentation in the game requires an emergent narrative to provide the flow through agents that reacts to the players’ actions and enacts significant aspect of the culture. We present here the existing agent modules and how they will be used in this project and the challenges in extending the work to this new domain.

Paper Nr: 264
Title:

Enhancing Clustering Technique with Knowledge-based System to Plan the Social Infrastructure Services

Authors:

Hesham A. Salman, Lamiaa Fattouh Ibrahim and Zaki Fayed

Abstract: This article present new algorithm for clustering data in the presence of obstacles. In real world, there exist many physical obstacles such as rivers, lakes, highways and mountains..., and their presence may affect the result of clustering significantly. In this paper, we study the problem of clustering in the presence of obstacles to solve location of public service facilities. Each facility must serve minimum pre-specified level of demand. The objective is to minimize the distance travelled by users to reach the facilities this means also to maximize the accessibility to facilities. To achieve this objective we developed CKB-WSP algorithm (Clustering using Knowledge-Based Systems and Weighted Short Path). This algorithm is Density-based clustering algorithm using Dijkstra algorithm to calculate obstructed short path distance where the clustering distance represents a weighted shortest path. The weights are associated with intersection node and represent the population number. Each type of social facility(schools, fire stations, hospitals, mosque, church…) own many constraints such as surface area and number of people to be served, maximum distance, available location to locate these services. All these constraints is stored in the Knowledge-Based system. Comparisons with other clustering methods are presented showing the advantages of the CKB-WSP algorithm introduced in this paper.

Posters
Paper Nr: 23
Title:

Artificial Intelligence and Creativity - Two Requirements to Solve an Extremely Complex Coloring Problem

Authors:

Bernd Steinbach and Christian Posthoff

Abstract: The topic of this paper is the rectangle-free coloring of grids using four colors which is equivalent to the edge coloring of complete bipartite graphs without complete monochromatic subgraphs K2,2. So far unsolved are the grids of the sizes 17×17, 17×18, 18×17, and 18×18. The number of different 4-color patterns of the grid 18×18 is equal to 4324 ≈ 1.16798∗10195. We summarize in this paper some basic approaches in order to gain the required knowledge. Three creative approaches are steps so solve the most complex grid of the size 18×18. Two advanced creative approaches reduce the required runtime to less than 12 percent.

Paper Nr: 59
Title:

An Enhanced Ant Colony Optimization for Routing Area Mobility Prediction over Cellular Communications Network

Authors:

Mohammad Sh. Daoud, Aladdin Ayesh, Mustafa Al-Fayoumi and Adrian A. Hopgood

Abstract: Cellular communication networks have become medium to provide various services. Most of the services provided are based on the users’ locations, as in location-based services (LBSs); these services include both common voice services as well as multimedia and integrated data services. Used techniques mostly suffered from complex computation, accuracy rate regression and insufficient accuracy. Nevertheless, in the cell side, reducing the complexity cost and preventing the prediction algorithm to perform in two closer time slot. That’s why using routing area should be able to avoid the cell side problems. This paper discusses An Enhanced Ant Colony Optimization for Routing Area Mobility Prediction over Cellular Communications Network (EACORA) which is based on developed ant colony Optimization.

Paper Nr: 60
Title:

Strategy Tree Construction and Optimization with Genetic Programming

Authors:

Chi Xu, Jianxiong Qiao and Na Jia

Abstract: We applied genetic programming (GP) to search for a strategy in a technical analysis (TA) indicator candidate pool for stock market trading and optimized it through historical data. The method provides decision rule optimization scheme to deal with problems in the real trading in financial market, and it optimizes strategies in relatively complicated contents. GP is used to construct the condition in decision rule with different logical operations. The method has been applied to the optimization of investment strategies with good return results in simulation experiments.

Paper Nr: 65
Title:

Ontology of Offers According to Ingarden’s Theory of Individual Objects

Authors:

Jan Andreasik

Abstract: In the paper, ontology of individual object and interpretation of it in the area of offers of goods and services is presented and reconstructed by the author. Ontology is defined on the basis of the Roman Ingarden’s formal ontology, especially, on the basis of theory of individual object.

Paper Nr: 72
Title:

Predicting Evacuation Capacity for Public Buildings

Authors:

Pejman Kamkarian and Henry Hexmoor

Abstract: This paper demonstrates a solution for analyzing public space evacuation rates. Evacuating from a public building in a reasonable amount of time is reliant upon how safe the space is in terms of achieving a minimum time to move people outside. In order to increase the safety of evacuation in public spaces, we employed the Bayesian Belief Network method. To have a better estimation pattern, we have to focus on important physical environmental features as well as crowd formation and specifications in a public space.

Paper Nr: 77
Title:

Evolutionary Systems Agents’ Mathematical Models

Authors:

Valentina N. Korzhova, Viktor V. Ivanov and Natalya V. Ivanova

Abstract: General mathematical theory of evolutionary system developed earlier is implemented to various problems of artificial intelligence and intelligent agent mathematical modeling. Examples of application of this general theory to the evolutionary systems such as economics, education, and health care are also considered.

Paper Nr: 85
Title:

The Effect of Mutation Operation on GP- based Stream Ciphers Design Algorithm

Authors:

Wasan Shakr Awad

Abstract: Mutation operation is used to introduce a small perturbation in the population from time to time so as to maintain its diversity. Several mutation operations have been developed for genetic programming. This paper is to study the impact of mutation operation on the performance of genetic programming. We present six types of mutation operations that have been applied in the simulated annealing programming (SAP) algorithm, which is an algorithm used to design stream ciphers using genetic programming and simulated annealing. Experiments performed to study the effectiveness of these operations in solving the underlying problem. It has been shown that mutation operation can affect the performance of genetic programming, especially when it is used to solve complex problems.

Paper Nr: 86
Title:

Using the Expanded IWO Algorithm to Solve the Traveling Salesman Problem

Authors:

Daniel Kostrzewa and Henryk Josiński

Abstract: The Invasive Weed Optimization algorithm (IWO) is an optimization metaheuristic inspired by dynamic growth of weeds colony. The authors of the present paper have expanded the strategy of the search space exploration of the IWO algorithm introducing a hybrid method along with a concept of the family selection applied in the phase of creating individuals. The goal of the project was to evaluate the expanded IWO version (exIWO) as well as the original IWO by testing their usefulness for solving some test instances of the traveling salesman problem (TSP) taken from the TSPLIB collection which allows to compare the experimental results with outcomes reported in the literature. The results produced by other heuristic algorithms as well as the methods based on the self-organizing maps served as the reference points.

Paper Nr: 94
Title:

A Novel Approach to MDO using an Adaptive Multi-Agent System

Authors:

Tom Jorquera, Jean-Pierre Georgé, Marie-Pierre Gleizes, Christine Régis and Pierre Glize

Abstract: MultiDisciplinary Optimization (MDO) problems represent one of the hardest and broadest domains of continuous optimization, often too complex to be tackled by classical optimization methods. We propose an original approach for taking into account this complexity using a self-adaptive multi-agent system where each elements of the problem become an agent in charge of a small part of the problem.

Paper Nr: 109
Title:

Multi-agent Reinforcement Learning based on Multi-channel ART Networks

Authors:

Hitomi Morishita, Hiroaki Ueda and Kenichi Takahashi

Abstract: 3-channel fuzzy ART network FALCON is a good solution to combine reinforcement learning with state segmentation, where it learns the relations among percepts, actions and rewards. FALCON, however, does not have a mechanism to predict behavior of other agents, and thus it is difficult for FALCON to learn the optimal agent’s behavior in a multi-agent circumstance. In this paper, an action prediction module based on 2-channel fuzzy ART network is proposed, and FALCON is modified in order to be able to register the output of the action prediction module. The modified FALCON is called FALCON AP. Moreover, FALCON ER that estimates the expected value of rewards and selects an action according to the value is proposed. Through experiments in which FALCON, FALCON AP and FALCON ER are applied to a card game Hearts, it is shown that FALCON ER receives less penalty points and learns better rules.

Paper Nr: 112
Title:

A Fuzzy Logic Model for Real-time Incident Detection in Urban Road Network

Authors:

Faisal Ahmed and Yaser E. Hawas

Abstract: Incident detection systems for the urban traffic network are still lacking efficient algorithms or models for better performance. This paper presents a new urban incident detection system based on the application of Fuzzy Logic modeling. Offline urban incident and corresponding non-incident scenarios are generated using a microscopic simulation model assuming varying traffic link flows, phase timing, cycle times, and link lengths. The traffic measures are extracted from three detectors on each link. Statistical significance analysis was utilized to identify the significant input variables to be used in developing the Neuro-fuzzy model. A set of data was generated and used for training of the proposed Neuro-fuzzy model, while another set was used for validation. The performance of the proposed model is assessed using the success and the false alarm rates of detecting an incident at a specific cycle time.

Paper Nr: 116
Title:

Data Mining Tool for Decision Support in Stock Market

Authors:

Sung-Dong Kim

Abstract: Stock investors want to make continuous profits in stock market. They have to choose profitable stocks and to follow the appropriate trading policy to achieve their goal. It is difficult for individual investors to determine what to buy and when to buy and sell. This paper proposes a data mining tool for stock investors’ decision support by recommending profitable stocks and proposing the trading policy. The proposed tool provides three functions: stock data management, stock price prediction model generation by applying the machine learning algorithms and the investment simulation for seeking the profitable trading policy. Users can generate and test the stock price prediction model by selecting their own technical indicators, simulate the trading and select the best trading policy through the evaluation of the trading results.

Paper Nr: 121
Title:

What Readers want to Experience - An Approach to Quantify Conversational Maxims with Preferences for Reading Behaviour

Authors:

Hanna Knäusl and Bernd Ludwig

Abstract: Searching information on web pages may be a tedious task for users as web pages are not tailored to the user’s current information need. This leaves an enormous workload for the user to filter the presented information and influences his emotional attitude towards the search task. In order to reduce the problem of information overload, we propose an approach to adapt the system’s response to the user’s reading experience preferences. Using these preferences, it is possible to select preferred content from a web page. We present a decision strategy for this selection of content on the basis of the learned preferences.

Paper Nr: 132
Title:

Cooperative Self-organization of Agents for Optimization - The Electrical Wiring Example

Authors:

Stéphanie Combettes, Pierre Glize, Thomas Sontheimer and Sylvain Rougemaille

Abstract: In aircrafts, densifying electrical systems and oversizing cables in order to respect constraints induce a useless increase in cable weight. This increase leads to additional costs of operation and to an unnecessary pollution during the plane operating life. In this paper we address optimization of harness weight which is a monoobjective problem with manifold and interdependent constraints. To solve this problem, we use a multi-agent approach based on the cooperative self-organization of agents. Performances obtained by the ’Smart Harness Optimizer’ software that we have developed are promising for problems considered by the experts as being very difficult. In this article, we expose the method used to solve this Constraint Optimization Problem. Then we apply it to the addressed problem and finally we give results on typical cases and analyze them.

Paper Nr: 135
Title:

Multi-agent Case-based Reasoning Inference Engine Proposal for Reusable Robotics

Authors:

Francisco Gindre, D. Kotlirevsky, S. Scaine, N. Jafelle Fraga and M. Daniela López De Luise

Abstract: Different articles and surveys on the matter identify a remarkable success of robotics on static environments that has not replicated on dynamic areas of application. This paper presents the proposal of a research project that identifies the main hurdles of this issue and present feasible solutions for the matter.

Paper Nr: 138
Title:

Description and Evaluation of Algorithms for Ontology Matching

Authors:

Mario Blanco-Alonso, Francisco J. Rodríguez-Martínez and Lorena Otero-Cerdeira

Abstract: In this paper, we present a state of the art about Ontology Matching Algorithms and we propose a general classification of them. A selection of three algorithms to work with a concrete platform is presented: CODI, LogMap and MaasMatch. In addition we propose a testbed divided in three groups of tests to evaluate the algorithms. These algorithms were tested and evaluated to verify which was the most suitable for this problem.

Paper Nr: 140
Title:

How Women Think Robots Perceive Them – as if Robots were Men

Authors:

Matthijs A. Pontier and Johan F. Hoorn

Abstract: In previous studies, we developed an empirical account of user engagement with software agents. We formalized this model, tested it for internal consistency, and implemented it into a series of software agents to have them build up an affective relationship with their users. In addition, we equipped the agents with a module for affective decision-making, as well as the capability to generate a series of emotions (e.g., joy and anger). As follow-up of a successful pilot study with real users, the current paper employs a non-naïve version of a Turing Test to compare an agent’s affective performance with that of a human. We compared the performance of an agent equipped with our cognitive model to the performance of a human that controlled the agent in a Wizard of Oz condition during a speed-dating experiment in which participants were told they were dealing with a robot in both conditions. Participants did not detect any differences between the two conditions in the emotions the agent experienced and in the way he supposedly perceived the participants. As is, our model can be used for designing believable virtual agents or humanoid robots on the surface level of emotion expression.

Paper Nr: 143
Title:

Hazard Avoidance Auto Control System for a Robotic Vehicle

Authors:

Jean-Luc Farrugia, Matthew Montebello and John Abela

Abstract: The adoption of technology in particular intelligent systems have been employed to assist and provide additional safety to drivers and passengers while travelling in cars. Reactive driving aids such as ESP and RSC activate when a problem is detected, whilst fully autonomous driving systems are especially prone to pitfalls of the automation paradox. In an effort to contribute to the welfare of drivers and passengers while potentially reduce the amount of traffic accidents, the research documented in this paper presents an intelligent robot control system which autonomously avoids hazards, including collisions and rollovers whilst retaining driver involvement, inspired by techniques in Aviation, and was designed so that it would scale-up to full-fledged vehicles and robots. The system employs fuzzy logic to seamlessly constrain and/or takes over the driver’s control in a collision-prone situation, whilst allowing the user to drive freely in open plains or perform evasive action. The control algorithm also performs manoeuvres similar to the ‘3-point-turn’ in situations when there is not enough space to steer away from collisions, whilst rollover protection successfully prevents the vehicle from exiting a safe driving envelope when steering. The results presented show that the system provides a robust performance in the face of variable communication delays, inaccurate sensing, crosstalk, and even deliberately hazardous user-input, adapting to unseen and varying environments, seamlessly blending driver and machine control.

Paper Nr: 150
Title:

Method of Enriching Queries in Process of Information Retrieval in Arabic

Authors:

Souheyl Mallat, Anis Zouaghi, Mounir Zrigui and Emma Hkiri

Abstract: In this paper, we propose a method of queries enrichment to improve the performance of information retrieval systems in Arabic. This method relies on the following steps: First step is the identification of significant terms (simple and composed) present in the query. The second consists in the generation of a descriptive list and its assignment to each term that has been identified as significant in the query. Final step is the application of the weighting functions of Salton TF-IDF and the TF-IEF on the list generated in the previous step. TF-IDF function identifies relevant documents, while the TF-IEF's role is to identify the relevant sentences, and to assign a weight to all the terms belonging to those sentences. The terms of high weight (which are terms that may be correlated to the context of the response) are incorporated into the original query.

Paper Nr: 159
Title:

Through a Fuzzy CTL Logic for Modelling Urban Trajectories - A Framework for Modelling City Evolution from Past to Future

Authors:

Asma Zoghlami, Cyril de Runz and Herman Akdag

Abstract: A city is by definition a relatively large town of a significant importance. It is a centre of population, commerce, culture, industry, etc. The city evolves over time and gets morphological, sociological, economic and political transformations. Geographic Information System (GIS) may be used in spatial analysis of both the current city and its evolution over time. Based on the past and the present of a city, we are interested in developing a methodology that goes from the spatiotemporal modelling of its evolution to its prediction in the future. The motivation behind this research is to create a tool for the decision support at the disposal of the town hall. This tool aims to help making future decisions about investments, transport networks, infrastructures, etc. In this paper, we propose a framework that allows defining the possible trajectories of the city following the spatial, temporal and functional dimensions. The definition of those trajectories will be attached to a reasoning based on logic according to modalities, time and the imperfect nature of the information (imprecision, uncertainty, etc.).

Paper Nr: 166
Title:

A Hybrid Intuitionistic MCDM Model for Supplier Selection

Authors:

Babak Daneshvar Rouyendegh

Abstract: This paper gives an overview of the Analytic Hierarchy Process (AHP) and Intuitionistic Fuzzy TOPSIS (IFT) methods. This study deals an evaluation methodology based on the AHP-IFT where the uncertainty is handeled with linguistic values. First, the supplier selection problem is formulated by AHP is used to determine weights of the criteria. In the second stage, IFT used to obtain full ranking among alternatives based on opinion of the Decision Makers (DMs). The present model provides an accurate and easy classification in supplier attributes by that have been prioritized in the hybrid model. A numerical example is given to clarify the main developed result in this paper.

Paper Nr: 171
Title:

A Study on Generation of Synthetic Evolving Social Graph

Authors:

Nagehan Ilhan and Şule Gündüz Öğüdücü

Abstract: Social networks are popular tools for communication, interaction, and information sharing on the Internet. The extreme popularity and rapid growth of these online social networks reveal to study, understand, and discover their properties. Social networks evolve gradually and the network structure varies as the network grows. Large-scale dynamic network analysis requires a large quantity of network data to be available for the experiments and using real data have restrictions due to the privacy issues. Synthetic data generation is an alternative way to overcome these problems. The challenge when generating synthetic data is having characteristics that are similar to real-world data. In this paper, we study on generating synthetic, but realistic, time-evolving social graphs. We describe two main classes of properties: static and dynamic. We analyzed real datasets and extracted their behavior using static and dynamic properties. Then, we generated synthetic graphs with different parameter settings using Barabasi-Albert model (Barabasi and Albert, 1999). Our work enables the creation of synthetic networks that reflect both static and dynamic characteristics of online social networks. Moreover, our generated data may lead to more accurate structural and growth models, which are useful for network analysis and planning.

Paper Nr: 182
Title:

Interactive Narration Requires Interaction and Emotion

Authors:

A. Pauchet, F. Rioult, E. Chanoni, Z. Ales and O. Şerban

Abstract: This paper shows how interaction is essential for storytelling with a child. A corpus of narrative dialogues between parents and their children was coded with a mentalist grid. The results of two modelling methods were analysed by an expert in parent-child dialogue analysis. The extraction of dialogue patterns reveals regularities explaining the character’s emotion. Results showed that the most efficient models contain at least one request for attention and/or emotion.

Paper Nr: 187
Title:

Numerical Kernels for Monitoring and Repairing Plans Involving Continuous and Consumable Resources

Authors:

Enrico Scala

Abstract: In this paper we introduce a technique for monitoring and repairing a plan dealing with continuous and consumable resources. The mechanism relies on the notion of numerical kernel. Concretely, a numerical kernel establishes the sufficient and necessary conditions for a plan to be valid in a specific state of the system. We employ the mechanism in a continual planning agent and we evaluate experimentally the approach for the Zenotravel domain. Results show good cpu-time w.r.t. a traditional replanning from scratch.

Paper Nr: 188
Title:

Game Theory and Learning at the Medium Access Control Layer for Distributed Radio Resource Sharing in Random Access Wireless Networks

Authors:

Eric Ayienga, Elisha Opiyo, Bernard Manderick and Okelo Odongo

Abstract: Game theory is not only useful in understanding the performance of human and autonomous game players, but it is also widely employed in solving resource allocation problems in distributed decision-making systems. Reinforcement learning is a promising technique that can be used by agents to learn and adapt their strategies in such systems. We have enhanced the carrier sense multiple access with collision avoidance mechanism used in random access networks by using concepts from the two fields so that nodes using different strategies can adapt to the current state of the wireless environment. Simulation results show that the enhanced mechanism outperforms the existing mechanism in terms of throughput, dropped packets and fairness. This is especially noticeable as the network size increases. However the existing mechanism performs better in terms of delay which can be attributed to increased processing.

Paper Nr: 207
Title:

An Event Metric and an Episode Metric for a Virtual Guide

Authors:

Felix Rabe and Ipke Wachsmuth

Abstract: In this paper we introduce a metric to compare events and episodes in the episodic memory system of a virtual agent. The agent, a virtual tour guide based on a belief – desire – intention cognitive architecture, uses his memories to improve the walks around a virtual city. The guide’s past experiences are memorized as events and organized into episodes. Each event is indexed along six dimensions and is comparable on each dimension with a distinct distance function. This is then utilized to measure the similarity between episodes.

Paper Nr: 211
Title:

Automated Planning for Pick-and-Place Robot

Authors:

Yazmin S. Villegas-Hernandez and Federico Guedea-Elizalde

Abstract: In this research was developed a language to describe a robot-based assembly. This language has an important role in the generation of robot programs. To accomplish with the objective of automatic generation of robot programs, it was developed a system, which consists on the next subsystems: a High-level-language Planner, a Generic-level-language Parser and a Wrapper-generic-level language.

Paper Nr: 225
Title:

An Argumentation System with Indirect Attacks

Authors:

Kazuko Takahashi

Abstract: We discuss argumentation frameworks with indirect attacks, such as why-questions and supports. A whyquestion is regarded as a kind of attack relation, and a support is an answer to an un-presented why-question. Based on this idea, we construct an argumentation framework with why-questions from a pair of knowledge bases, as an instantiation of Dung’s abstract argumentation framework, and show that its extension is consistent. Next, we transform this argumentation framework into an argumentation framework with supports, and discuss its properties. The resulting framework is an instantiation of Bipolar Argumentation Framework (BAF), defined as a triple consisting of arguments, attack relations and support relations. We define an extension of BAF, and show that the framework defined in this paper has some nice properties.

Paper Nr: 231
Title:

A Study of Decision-making Model Considering Priorities based on Two Kinds of Evaluation - Decision Making Methodology Applying Risk Evaluation based on Prospect Theory

Authors:

Rumiko Azuma and Shinya Nozaki

Abstract: The Analytic Hierarchy Process (AHP) is a multi-criteria decision-making approach aimed at reflecting a human’s subjective judgment or vagueness. The conventional evaluation in AHP is considered to be a kind of utility. However, there are some cases where the traditional utility theory cannot explain risk aversion. This paper presents a new decision-making methodology for considering risk evaluation. We propose the hierarchy model that contains return and risk categories, and an AHP method that applies prospect theory, which is able to explain people’s decisions when they face situations involving risks. Therefore, by proposing an AHP method that utilizes it, we enable the evaluation of alternatives under return and risk.

Paper Nr: 235
Title:

The Characterisation and Optimisation of TLC NAND Flash Memory using Machine Learning - A Position Paper

Authors:

Sorcha Bennett and Joe Sullivan

Abstract: Flash memory is non-volatile and, while it is becoming ever more commonplace, it is not yet a complete replacement for hard disk drives. The physical layout of Flash means that it is more susceptible to degradation over time, leading to a limited lifetime of use. This paper will give an introduction to NAND Flash memory, followed by an overview of the relevant research on the reliability of MLC memory, conducted using Machine Learning (ML). The results obtained will then be used to characterise and optimise the reliability of TLC memory.

Paper Nr: 236
Title:

Peer-to-Peer MapReduce Platform

Authors:

Darhan Ahmed-Zaki, Grzegorz Dobrowolski and Bolatzhan Kumalakov

Abstract: Publicly available Peer-to-PeerMapReduce (P2P-MapReduce) frameworks suffer lack of practical implementation, which significantly reduces their role in system engineering. Presented research supplies valuable data on working implementation of P2P-MapReduce platform. Resulting novelties include advanced workload distribution function, which integrates mobile devices as execution nodes; novel computing node and multi-agent system architectures.

Paper Nr: 241
Title:

An Evolutionary View of Collective Intelligence

Authors:

Bengt Carlsson and Andreas Jacobsson

Abstract: Based on the question “How can people and computers be connected so that – collectively – they act more intelligently than any individuals, groups, or computers have ever done before?” we propose an evolutionary approach. From this point of view, there are of course fundamental differences between man and machine. Where one is artificial, the other is natural, and where the computer needs to process, the brain must adapt. We propose the use of culturally inherited units, i.e., memes, for describing collective knowledge storage. Like the genes, memes have the ability to be inherited to the next generation. Genes appear independently of our society while memes are a result of our cultural development. The concept of collective intelligence may involve a new kind of meme, entirely emerging within the intersection between man and machine, i.e., outside the scope of human control. The challenge is to model this behavior without overriding constraints within basic evolutionary vs. machine settings.

Paper Nr: 245
Title:

A Recommendation Engine for Subsistence Farmers

Authors:

Daniel P. Stormont and Samantha L. Snabes

Abstract: Many of the world’s agricultural workers, especially in the equatorial regions, are subsistence farmers. This paper describes the design of a Recommendation Engine that is the core of a crowd-sourced tool to assist subsistence farmers in better utilizing their land. This tool is being developed by a worldwide team of developers through hackathons and volunteer work. The paper provides an introduction to the project, an overview of the system design, current and future designs planned for the Recommendation Engine, and the next steps for the project.

Paper Nr: 255
Title:

Automatic Face Corpus Creation

Authors:

Ladislav Lenc and Pavel Král

Abstract: This paper deals with the automatic real-world face corpus creation. The main contribution consists in proposition and evaluation of the automatic face corpus creation algorithm. Next, we statistically analysed the structure of the created face corpus when the automatic algorithm is used. We further compared the face recognition accuracy of our previously developed face recognition approach on this corpus while using different size/quality datasets. We have shown that the manual verification of the corpus is not necessary. Therefore, we concluded that our proposed algorithm is suitable for the further use by the Czech News Agency, our commercial partner.

Paper Nr: 257
Title:

From 7 to 77 - I Teach You and You Teach Me!

Authors:

Tiago Amaral and Isabel Machado Alexandre

Abstract: This paper presents a project that aims to devise an innovative interactive application, more precisely a serious game, covering the topic of financial education. The topic is quite relevant nowadays, not only in Europe but worldwide, and is also one of the major issues on most government’s agendas. The aim of this application is by using a storytelling framework devise a collaborative platform, where people from 7 to 77 acquire or enhance their financial skills. To do this, the application tackles several research fields such as interactive digital storytelling, collaboration human-computer interaction, autonomous agents, etc.

Paper Nr: 259
Title:

Intelligent Predicting Method of Water Bloom based RBFNN and LSSVM

Authors:

Liu Zaiwen, Wu Qiaowei and Lv Siying

Abstract: Water bloom is one phenomena of eutrophication, and water bloom prediction is always a challenge. A short-term intelligent predicting method based on RBF neural network (RBFNN), and medium-term intelligent predicting method based on least squares support vector machine (LSSVM) for water bloom are proposed in this paper. Including research on the monitoring learning algorithms to the center, width and weight of basis function of RBF network, the width of RBF and fitting and generalization abilities of network, and the function and influence, which the number of RBF hidden level nodes brings to the performance of network, as well as error-corrected algorithm based on gradient descent are analyzed. Least squares support machine, which has long prediction period and high degree of prediction accuracy, needs a small amount of sample can be used to predict the medium-term change discipline of Chl-a (Chlorophyll-a) well. The results of simulation and application show that: RBF neural network can be used to forecast the change of Chl-a in short term well, and LSSVM improves the algorithm of support vector machine (SVM), and it has long-term prediction period, strong generalization ability and high prediction accuracy; and this model provides an efficient new way for medium-term water bloom prediction.

Area 2 - Agents

Full Papers
Paper Nr: 16
Title:

Synchronizing for Performance - DCOP Algorithms

Authors:

Or Peri and Amnon Meisels

Abstract: The last decade has given rise to a large variety of search algorithms for distributed constraints optimization problems (DCOPs). All of these distributed algorithms operate among agents in an asynchronous environment. The present paper presents a categorization of DCOP algorithms into several classes of synchronization. Algorithms of different classes of synchronization are shown to behave differently with respect to idle time of agents and to irrelevant computation. To enable the investigation of the relation between the classes of synchronization of algorithms and their run-time performance, one can control the asynchronous behavior of the multi-agent system by changing the amount of message delays. A preliminary probabilistic model for computing the expected performance of DCOP algorithms of different synchronization classes is presented. These expectations are realized in experiments on delayed message asynchronous systems. It turns out that the performance of algorithms of a weaker synchronization class deteriorates much more when the system becomes asynchronous than the performance of more synchronized DCOP algorithms. The notable exception is that concurrent algorithms, that run multiple search processes, are much more robust to message delays than all other DCOP algorithms.

Paper Nr: 29
Title:

Distributed Envy Minimization for Resource Allocation

Authors:

Arnon Netzer and Amnon Meisels

Abstract: The allocation of indivisible resources to multiple agents generates envy among the agents. An Envy Free allocation may not exist in general and one can search for a minimal envy allocation. The present paper proposes a formulation of this problem in a distributed search framework. Distributed Envy Minimization (DEM) - A Branch and Bound based distributed search algorithm for finding the envy minimizing allocation is presented and its correctness is proven. Two improvements to the DEM algorithm are presented - Forward Estimate (DEM-FE) and Forward Bound (DEM-FB). An experimental evaluation of the three algorithms demonstrates the benefit of using the Forward Estimate and Forward Bound techniques.

Paper Nr: 93
Title:

Evaluation of a Self-organizing Heuristic for Interdependent Distributed Search Spaces

Authors:

Christian Hinrichs, Michael Sonnenschein and Sebastian Lehnhoff

Abstract: Whenever multiple stakeholders try to optimize a common objective function in a distributed way, an adroit coordination mechanism is necessary. This contribution presents a formal model of distributed combinatorial optimization problems. Subsequently, a heuristic is introduced, that uses self-organizing mechanisms to optimize a common global objective as well as individual local objectives in a fully decentralized manner. This heuristic, COHDA2, is implemented in an asynchronous multi-agent system, and is being extensively evaluated by means of a real-world problem from the smart grid domain. We give insight into the convergence process and show the robustness of COHDA2 against unsteady communication networks. We show that COHDA2 is a very efficient decentralized heuristic that is able to tackle a distributed combinatorial optimization problem with regard to multiple local objective functions, as well as a common global objective function, without being dependent on centrally gathered knowledge.

Paper Nr: 97
Title:

Hierarchical Roadmap Approach to Rough Terrain Motion Planning

Authors:

Michael Brunner, Bernd Brüggemann and Dirk Schulz

Abstract: A reconfigurable chassis provides a mobile robot with a high degree of mobility and enables it to overcome rough terrain in unstructured outdoor environments, like boulders or rubble, and challenging structures in urban environments, like stairs or steps. Yet, many planning algorithms rarely exploit those enhanced capabilities to the full extent, limiting these systems to mainly flat environments also traversable by less capable fixed-chassis robots. In this paper we introduce a two-stage roadmap approach to motion planning for reconfigurable robots which utilizes the system's actuators to traverse rough terrain and obstacles. First, by considering the platform's operating limits rather than the complete state, we quickly generate an initial path. Second, we refine the initial path in rough areas within a constrained search space. So we are able to plan appropriate actuator configurations to traverse rough areas and ensure the system's safety. Our algorithm does not categorize the terrain and does not use any predefined motion sequences. Hence, our planner can be applied to urban structures, like stairs, as well as rough unstructured environments. We present simulation experiments to provide more insight into our method and real-world experiments to prove the feasibility of our motion planning approach on a real robot.

Paper Nr: 100
Title:

Studying Aviation Incidents by Agent-based Simulation and Analysis - A Case Study on a Runway Incursion Incident

Authors:

Tibor Bosse and Nataliya M. Mogles

Abstract: This paper introduces an agent-based approach to analyse the dynamics of accidents and incidents in aviation. The approach makes use of agent-based simulation on the one hand, and of formal verification of dynamic properties on the other hand. The simulation part enables the analyst to explore various hypothetical scenarios under different circumstances, with an emphasis on error related to human factors. The formal verification part enables the analyst to identify scenarios involving potential hazards, and to relate those hazards (via so-called interlevel relations) to inadequate behaviour on the level of individual agents. The approach is illustrated by means of a case study on a runway incursion incident, and a number of advantages with respect to the current state-of-the-art are discussed.

Paper Nr: 105
Title:

Evolution of Cooperation in Packet Forwarding with the Random Waypoint Model

Authors:

Jeffrey Hudack, Nathaniel Gemelli and Jae Oh

Abstract: In multi-agent systems with self-interested individuals interacting locally, it can be difficult to determine if cooperative behavior will emerge. Evolutionary Game Theory provides some valuable tools to this end, but is not suited to systems with dynamic models of interaction. Mobile ad hoc networks provide a compelling application for evolutionary game theory, but there are still significant gaps between the theoretical results and the practical challenges. We discuss and provide some of the assumptions necessary to apply previous work in evolutionary game theory to the ad hoc network packet routing domain. We then analyze the similarities and differences between Brownian mobility and Random Waypoint mobility and show that convergence to cooperation requires a significant reduction in velocity for the Random Waypoint model. Our contribution is to provide evidence that more realistic mobility models can make convergence to cooperation more difficult than previously shown using random methods.

Paper Nr: 125
Title:

Multiagent Model to Reduce the Bullwhip Effect

Authors:

Borja Ponte and David de la Fuente

Abstract: There are several circumstances which, in recent decades, have granted the supply chain management a strategic role in the search for competitive advantage. One of the goals is, undoubtedly, the reduction of Bullwhip Effect, which is generated by the amplification of the variability of orders along the chain, from the customer to the factory. This paper applies multiagent methodology for reducing Bullwhip Effect. To do this, it considers the supply chain as a global multiagent system, formed in turn by four multiagent subsystems. Each one of them represents one of the four levels of the traditional supply chain (Shop Retailer, Retailer, Wholesaler and Factory), and it coordinates various intelligent agents with different objectives. Thus, each level has its own capacity of decision and it seeks to optimize the supply chain management. The problem is analyzed both from a non collaborative approach, where each level seeks the optimal forecasting methodology independently of the rest, and from a collaborative approach, where each level negotiates with the rest looking for the best solution for the whole supply chain.

Paper Nr: 128
Title:

Exploring Assignment-Adaptive (ASAD) Trading Agents in Financial Market Experiments

Authors:

Steve Stotter, John Cartlidge and Dave Cliff

Abstract: Automated trading systems in the global financial markets are increasingly being deployed to do jobs previously done by skilled human traders: very often a human trader in the markets simply cannot tell whether the counter-party to a trade is another human, or a machine. Clearly, automated trading systems can easily be considered as “intelligent” software agents. In this paper we report on experiments with software traderagents running the well-known “AA” and “ZIP” strategies, often used as reference benchmarks in previously published studies; here we suggest disambiguated standard implementations of these algorithms. Then, using Exchange Portal (ExPo), an open-source financial exchange simulation platform designed for real-time behavioural economic experiments involving human traders and/or trader-agents, we explore the impact of introducing a new method for assignment adaptation in ZIP. Results show that markets containing only assignmentadaptive (ASAD) agents equilibrate more quickly after market shocks than markets containing only “standard” ZIP agents. However, perhaps counter-intuitively, in mixed heterogeneous populations of ASAD agents and ZIP agents, ZIP agents outperform ASAD agents. Evidence suggests that the behaviour of ASAD agents act as a new signal in the market that ZIP agents then use to beneficially alter their own behaviour, to the detriment of the ASAD agents themselves.

Paper Nr: 130
Title:

Managing Personality Influences in Dialogical Agents

Authors:

Jean-Paul Sansonnet and François Bouchet

Abstract: We present in this article an architecture implementing personality traits from the FFM/NEO PI-R taxonomy as influence operators upon the rational decision making process of dialogical agents. The objective is to separate designer-dependent resources (traits taxonomies, influence operators, behaviors/operators links) from the core part of the computational implementation (the personality engine). Through a case study, we show how our approach makes it easier to combine various resources and to observe various scenarios within a single framework.

Paper Nr: 161
Title:

Multi-player Multi-issue Negotiation with Mediator using CP-nets

Authors:

Thiri Haymar Kyaw, Sujata Ghosh and Rineke Verbrugge

Abstract: This paper presents a simple interactive negotiation approach for conflicts in everyday life with incomplete information. We focus on mediation to obtain an agreement while going through alternating offers over a finite time bargaining game. The mediator searches and proposes a jointly optimal negotiation text for all players participating in the negotiation process based on their conditional preference networks (CP-nets). The players make a decision whether to accept or reject by examining their utility CP-nets. We develop two algorithms for the mediator and the players. If the first negotiation text cannot be accepted by all players, the mediator offers the next negotiation texts by searching for jointly optimal solutions. This negotiation process continues until an agreement is achieved or a deadline is reached. This proposed approach can support multi-issue, multi-party negotiation to achieve an agreement during a finite number of rounds with near optimal outcomes.

Paper Nr: 192
Title:

Norm-regulated Transition System Situations

Authors:

Magnus Hjelmblom

Abstract: Many multi-agent systems (MAS) and other kinds of dynamic systems may be modeled as transition systems, in which actions are associated with transitions between di¤erent system states. This paper presents an approach to normative systems in this context, in which the permission or prohibition of actions is related to the permission or prohibition of di¤erent types of state transitions with respect to some condition d on a number of agents in a state. It introduces the notion of a norm-regulated transition system situation, which is intended to represent a single step in the run of a (norm-regulated) transition system. The normative framework uses an algebraic representation of conditional norms and is based on a systematic exploration of the possible types of state transitions with respect to d. A general-level Java/Prolog framework for norm-regulated transition system situations is currently being developed.

Paper Nr: 195
Title:

ERAM - Evacuation Routing using Ant Colony Optimization over Mobile Ad Hoc Networks

Authors:

Alejandro Avilés del Moral, Munehiro Takimoto and Yasushi Kambayashi

Abstract: This paper proposes a distributed multi-agent framework for discovering and optimizing evacuation routes on demand. We name it Evacuation Routing using Ant Colony Optimization over Mobile Ad hoc Networks (ERAM). Taking advantage of ant colony optimization (ACO) on mobile ad hoc networks (MANETs) composed of smartphones with geo-location capabilities, ERAM aims for adaptability and layout independence, relying exclusively on crowd’s knowledge during mass evacuations. Such knowledge is inserted into the system: actively, by users’ indication of having reached safe areas on their smartphones; and passively, by smartphones tracking their own movement. In the framework, agents migrate through nodes of a MANET towards safe areas based on an indirect communication mechanism called stigmergy, which is a behaviour that social insects show. Once an agent finds such an area, it traces its path backwardly collecting geographical information of intermediate nodes for composing an evacuation route. During the backward travel, agents lay pheromone down while they migrate back based on the ACO algorithm, strengthening quasi-optimal physical routes, and hence guiding succeeding agents. This scenario is analogous to data-packet routing on Internet or resource discovery on P2P networks, except it routes people through physical environments towards safe areas instead.

Paper Nr: 214
Title:

Planning Practical Paths for Tentacle Robots

Authors:

Jing Yang, Robert Codd-Downey, Patrick Dymond, Junquan Xu and Michael Jenkin

Abstract: Robots with many degrees of freedom with one fixed end are known as tentacle robots due to their similarity to the tentacles found on squid and octopus. Tentacle robots offer advantages over traditional robots in many scenarios due to their enhanced flexibility and reachability. Planning practical paths for these devices is challenging due to their high degrees of freedom (DOFs). Sampling-based path planners are a commonly used approach for high DOF planning problems but the solutions found using such planners are often not practical in that they do not take into account soft application-specific constraints during the planning process. This paper describes a general sample adjustment method for tentacle robots, which adjusts the randomly generated nodes within their local neighborhood to satisfy soft constraints required by the problem. The approach is demonstrated on a planar tentacle robot composed of ten Robotis Dynamixel AX-12 servos.

Short Papers
Paper Nr: 10
Title:

Associative Reinforcement Learning - A Proposal to Build Truly Adaptive Agents and Multi-agent Systems

Authors:

Eduardo Alonso and Esther Mondragón

Abstract: In this position paper we propose to enhance learning algorithms, reinforcement learning in particular, for agents and for multi-agent systems, with the introduction of concepts and mechanisms borrowed from associative learning theory. It is argued that existing algorithms are limited in that they adopt a very restricted view of what “learning” is, partly due to the constraints imposed by the Markov assumption upon which they are built. Interestingly, psychological theories of associative learning account for a wide range of social behaviours, making it an ideal framework to model learning in single agent scenarios as well as in multi-agent domains.

Paper Nr: 11
Title:

Soft Control of Swarms - Analytical Approach

Authors:

Guillaume Sartoretti and Max-Olivier Hongler

Abstract: We analytically study the collective dynamics of mutually interacting heterogeneous agents evolving in a random environment. Our formal framework consists of a collection of N scalar drifted Brownian motions (BM) diffusing on R. The mutual interactions are introduced via a ranked-based, real-time mechanism always endowing the laggard (i.e the agent with the leftmost position) with an extra positive drift. The extra drift generates a net tendency for any agents not to remain the laggard of the society. For well chosen individual and extra laggard’s drifts, the agents organize with time to flock towards a tight and stable travelling spatial pattern. For a population of (N −1) identical agents and an atypical fellow (called hereafter the shill), we are able to analytically discuss the dynamics. In particular we exhibit how a single turbulent shill, stylized here by a ballistic diffusion process, can destroy the cohesion of a swarm. Conversely, we also analytically show how a single shill is able to safely pilot a whole swarm to avoid an obstacle, via interactions with its fellows. A series of simulations experiments comfort our analytic findings.

Paper Nr: 27
Title:

An Agent-based Framework for Intelligent Optimization of Interactive Visualizations

Authors:

Pedro Miguel Moreira, Luis Paulo Reis and A. Augusto Sousa

Abstract: Interactive visualization of virtual environments is an active research topic. There is a multiplicity of applications such as simulation systems, augmented and mixed reality environments, computer games, amongst others, which endlessly demand for greater levels of realism and interaction. At every stage of the process, including modeling, image synthesis, transmission and navigation, there are identifiable circumstances which may compromise the achievement of high quality solutions for the posed problems. For many of these problems, an effective use of optimization tools can play a major role in order to achieve solutions with better quality. Within this context, an innovative optimization architecture is presented regarding to two major principles. The first principle comprises the possibility to integrate, with reduced effort, the optimization tools with existent applications and systems. Thus, we propose an agent-based framework where the optimization application may operate as an independent process in respect to the visualization application where communication is achieved by means of a specifically developed high-level message based protocol. The second principle establishes on the utilization of a class of intelligent optimization methods, known as metaheuristics, which major distinguishing quality is their great level of problem-independence, thus, enabling a wider application. The paper describes conducted experiments and presents results that demonstrate the utility and efficacy of the proposed framework.

Paper Nr: 33
Title:

Multiparty Argumentation Game for Consensual Expansion

Authors:

Stefano Bromuri and Maxime Morge

Abstract: We consider here a set of agents, each of them having her own argumentation. Arguments and conflicts between them are subjective. The aim of each agent is to enrich her argumentation by taking into account the arguments and conflicts of the other agents. We adopt here an individual-based approach where the crossfertilization of argumentations emerge from the interactions between the agents. For this purpose, we formalize a multi-party argumentation game using Event Calculus. At the end of the game, each agent extends its argumentation by using the arguments exchanged and the conflicts shared. As we show formally, such an expansion is consensual. By adopting an individual-based approach, our model is explanatory since it highlights the conflicts between the agents.

Paper Nr: 36
Title:

Agent-based Modeling of Indoor Evacuation Behavior under Stressful Psychological State

Authors:

Lu Tan and Hui Lin

Abstract: This paper presents an agent-based model focusing on occupant’s locomotion under stress, so as to study the impact of psychic stress on evacuation efficiency. In our model, the occupant’s stress is determined by factors including the moving velocity and distance to the exit, the psychological feature of stress resistance capability, and emotional contagion. The occupant’s evacuation behaviour in a stressful psychological state is attained as an emergent function of stress-related desire intensity and interaction force based on the Helbing social force model. Through a series of simulations using the proposed model, it is concluded that the increase in the occupants’ stress level does reduce the evacuation efficiency; the emotional contagion performs either a panic effect or a calm-down effect which affects the evacuation efficiency negatively or positively; intensive emotional contagion will significantly affect the occupant’s stress level and lead to remarkable variation in evacuation time; population structure has an influence on the evacuation efficiency in respect to the occupant’s capability of stress resistance. The conclusions indicate that proper control of psychic stress during emergency evacuation is critical for improving evacuation efficiency.

Paper Nr: 37
Title:

A Prospect on How to Find the Polarity of a Financial News by Keeping an Objective Standpoint - Position Paper

Authors:

Roxana Bersan, Dimitrios Kampas and Christoph Schommer

Abstract: This position paper raises the question on how we can keep an independent standpoint regarding the finding of a polarity in a news document. As we know, an usefulness and relevance of a text news may be seen differently by a group of evaluators. The differences are depending on their interests, their knowledge, and/or their ability to understand. Recent research in literature mostly follow a top-down approach, which is either a context-based solution or a dictionary-based approach. With respect to the perspective (standpoint) of an evaluator, we therefore come up with an alternative approach, which is bottom-up and which tends to overcome the power of a single evaluator. The idea is to introduce a collection of theme-related artificial agents (financial, economic, or political, . . . ), which are able to vote. A decision regarding the polarity of a financial news bases on the interplay of a social collection of agents (a swarm), which serve and assist the artificial agents while fulfilling simple (linguistic, statistical) tasks.

Paper Nr: 44
Title:

Activation of the Following Mode to Simulate Heterogeneous Pedestrian Behavior in Crowded Environment

Authors:

Laure Bourgois, Thomas Heckmann, Emmanuelle Grislin-Le Strugeon and Jean-Michel Auberlet

Abstract: To simulate pedestrian crowds, most of the current studies use the microscopic approach, in which the pedes- trian is modeled as an individual entity. With the microscopic approach, the heterogeneity in the pedestrian population is mostly based on inter-individual difference in the agent model parameters, like speed, destina- tion, etc. In fact, what can be seen in congested real situations, is some pedestrians choosing to temporarily follow other ones in order to facilitate the flow while going on avoiding collisions. Each pedestrian can choose to adopt and leave such a behavior according to his/her individual and local situation. In order to model and simulate this behavior, we propose to include in the pedestrian agent navigational model a decision process that allows to activate the following task in addition to the collision avoidance task. The decision depends on the agent’s internal state and on what it perceives in its environment. Our proposition is an attempt to define the conditions that must trigger the switching to another navigation mode, and the selection of the other agent to be followed, i.e. the “leader”.

Paper Nr: 74
Title:

CPNT-Jade Framework - Developing Agent-based Control of Discrete Event Systems

Authors:

Andrzej Bożek and Marian Wysocki

Abstract: A framework built as a combination of CPN Tools software and JADE (Java Agent DEvelopment Frame-work) platform has been proposed. CPN Tools makes it possible to model a controlled system with the use of hierarchical timed colored Petri nets (HTCPN), a powerful formalism for modeling complex discrete event systems. The framework supports real-time mode of simulation of the target system model. Issues of communication and time synchronization between the two software components are transparent to a user. An example of the framework use has been presented. It refers to a dynamic flexible job shop manufacturing system for which an agent-based control has been developed.

Paper Nr: 80
Title:

Optimizing Theta Model for Monthly Data

Authors:

Fotios Petropoulos and Konstantinos Nikolopoulos

Abstract: Forecasting accuracy and performance of extrapolation techniques has always been of major importance for both researchers and practitioners. Towards this direction, many forecasting competitions have conducted over the years, in order to provide solid performance measurement frameworks for new methods. The Theta model outperformed all other participants during the largest up-to-date competition (M3-competition). The model’s performance is based to the a-priori decomposition of the original series into two separate lines, which contain specific amount of information regarding the short-term and long-term behavior of the data. The current research investigates possible modifications on the original Theta model, aiming to the development of an optimized version of the model specifically for the monthly data. The proposed adjustments refer to better estimation of the seasonal component, extension of the decomposition feature of the original model and better optimization procedures for the smoothing parameter. The optimized model was tested for its efficiency in a large data set containing more than 20,000 empirical series, displaying improved performance ability when monthly data are considered.

Paper Nr: 99
Title:

Experimental Evaluation of the Effects of Manipulation by Merging in Weighted Voting Games

Authors:

Ramoni O. Lasisi and Vicki H. Allan

Abstract: This paper considers weighted voting games and a method of manipulating those games, called merging. This manipulation involves a coordinated action among some agents who come together to form a bloc by merging their weights in order to have more power over the outcomes of the games. We conduct careful experimental investigations to evaluate the opportunities for beneficial merging available for strategic agents using two prominents power indices: Shapley-Shubik and Banzhaf indices. Previous work has shown that finding a beneficial merge is NP-hard for both the Shapley-Shubik and Banzhaf indices, and leaves us with the impression that this is indeed so in practice. However, results from our experiments suggest that finding beneficial merge is relatively easy in practice. Furthermore, while it appears that we may be powerless to stop manipulation by merging for a given game, we suggest a measure, termed quota ratio, that the game designer may be able to control. Thus, we deduce that if we know that the quota ratio of a game is high, we would feel more comfortable about the honesty of the game as the percentage of beneficial merges reduces. Finally, we conclude that the Banzhaf index may be more desirable to avoid manipulation by merging, especially for high values of quota ratios.

Paper Nr: 107
Title:

Smart Household - Selected Problem Solutions using Intelligent Controllable Electric Appliances

Authors:

Miroslav Prýmek and Aleš Horák

Abstract: In the current perspectives, renewable power sources bring new challenges for the power distribution. Substantial advances in the reliability and flexibility in the overall power consumption can be achieved via a network of intelligent and controllable appliances, especially on the micro level, i.e. on the level of individual electric appliances within the scope of one household or one institution. In the paper, we identify the typical problems of the smart household approach and present a communication and control model which offers a solution to these problems based on the multi-agent system approach.

Paper Nr: 115
Title:

Creation of Creative Work Teams using Multi-Agent based Social Simulation

Authors:

Adrián Bresó, Alfonso Pérez, Javier Juan-Albarracín, Juan Martínez-Miranda, Montserrat Robles and Juan Miguel García-Gómez

Abstract: Over the past decades, advances in Artificial Intelligence (AI) techniques have investigated the modelling of complex systems. In particular, the use of Multi-Agent Systems (MAS) opened new possibilities for studying different domains using social simulation. In the present work we have implemented and empirically evaluated a Multi-Agent Based Social Simulation (MABSS) system to support the formation of creative work teams. Based on existent psychological and organizational creativity studies, we have modelled a set of personal characteristics and contextual factors to represent and analyse their influence on creativity at both: the individual and the group level. The obtained initial results were significantly better than the results obtained with a pure stochastic model (average improvement of 8.2%). Additionally, we empirically confirm some hypothesis about group formation from the organizational studies

Paper Nr: 117
Title:

Suppressing Energy Consumption of Transportation Robots using Mobile Agents

Authors:

Ryosuke Shibuya, Munehiro Takimoto and Yasushi Kambayashi

Abstract: This paper presents an application for controlling multiple robots connected by communication networks. Instead of making multiple robots pursue several tasks simultaneously, the framework makes mobile software agents migrate from one robot to another to perform the tasks. Since mobile software agents can migrate to arbitrary robots by wireless communication networks, they can find the most suitably equipped and/or the most suitably located robots to perform their task. We previously implemented an application of searching targets in our framework, and showed that it could suppress energy consumption of the entire system. In this paper, we propose a new mobile agent system that transports the found targets to a collection area. The approach is based on passing the targets among robots through throwing, and therefore, suppresses energy consumption as the same manner for searching targets. In order to show the effectiveness of our approach, we have implemented two strategies on a simulator and conducted numerical experiments. As a result, we show that our approach has more advantages than previous ones, and there is a remarkable difference between the two strategies in terms of energy saving.

Paper Nr: 141
Title:

SCENARIO - Setting Crowd Events using Augmented Reality and Artificial Intelligence

Authors:

Vincent-Emmanuel Farrugia, Matthew Montebello and Alexiei Dingli

Abstract: Crowd events pose numerous challenges on organisers, security personnel, and attendees who experience the event and who potentially can be at risk should an emergency arise. The research documented in this paper presents a system, SCENARIO, which aims to provide tools to aid organisers and attendees during crowd events. The system automates the organisation of the layout of attractions, attempts to reduce crowd congestion and enhances attendees’ safe experience. An optimiser agent was developed to solve a variant of the NP-Complete Facility Layout Problem to reduce congestion for venues which may include obstructions. Furthermore, an agent-based simulator is provided for visualising the effect of a layout on crowd flows, with the employment of a combination of a number of path-finding techniques. Moreover, a mobile agent, deployed on a hand-held device, visually and dynamically lays out an optimised path for attendees to assist their transition to their chosen attraction making use of path-finding structures, GPS and Augmented Reality, together with on-board sensors. The results presented show that the time for the layout optimisation convergence varies depending on complexity, with satisfactory layout output. Simulations run at efficient rates and agents keep their mobile state and avoid extreme congestion.

Paper Nr: 146
Title:

Planning and Reactive Agents in Dynamic Game Environments - An Experimental Study

Authors:

Roman Barták, Cyril Brom, Martin Černý and Jakub Gemrot

Abstract: Many contemporary computer games can be described as dynamic real-time simulations inhabited by autonomous intelligent virtual agents (IVAs) where most of the environmental structure is immutable and navigating through the environment is one of the most common activities. Though controlling the behaviour of such agents seems perfectly suited for action planning techniques, planning is not widely adopted in existing games. This paper contributes to discussion whether the current academic planning technology is ready for integration to existing games and under which conditions. The paper compares reactive techniques to classical planning in handling the action selection problem for IVAs in game-like environments. Several existing classical planners that occupied top positions in the International Planning Competition were connected to the virtual environment of Unreal Development Kit via the Pogamut platform. Performance of IVAs employing those planners and IVAs with reactive architecture was measured on a class of game-inspired maze-like test environments under different levels of external interference. It was shown that agents employing classical planning techniques outperform reactive agents if the size of the planning problem is small or if the environment changes are either hostile to the agent or not very frequent.

Paper Nr: 151
Title:

Towards Computational Models for a Long-term Interaction with an Artificial Conversational Companion

Authors:

Sviatlana Danilava, Stephan Busemann, Christoph Schommer and Gudrun Ziegler

Abstract: In this paper we describe a design approach for an Artificial Conversational Companion according to earlier identified requirements of utility, adaptivity, conversational capabilities and long-term interaction. The Companion is aimed to help advanced learners of a foreign language to practice conversation via instant messenger dialogues. In order to model a meaningful long-term interaction with an Artificial Conversational Companion for this application case, it is necessary to understand how natural long-term interaction via chat between human language experts and language learners works. For this purpose, we created a corpus from instant messenger-based interactions between native speakers of German and advanced learners of German as a foreign language. We used methods from conversation analysis to identify rules of interaction. Examples from our data set are used to illustrate how particular requirements for the agent can be fulfilled. Finally, we outline how the identified patterns of interaction can be used for the design of an Artificial Conversational Companion.

Paper Nr: 160
Title:

Optimal Decision Making in Agent-based Autonomous Groupage Traffic

Authors:

Stefan Edelkamp and Max Gath

Abstract: The dynamics and complexity of planning and scheduling processes in groupage traffic require efficient, proactive, and reactive system behavior to improve the service quality while ensuring time and cost efficient transportation. We implemented a multiagent-system to emerge an adequate system behavior and focus on the decision making processes of agents that is based on the Traveling Salesman Problem (TSP) with aspects like contract time windows, individual restricted capacities of trucks, premium services and varying priorities of dynamically incoming orders. We present an optimal depth-first branch-and-bound asymmetric TSP solver with constraints on tour feasibility and depot reachability at any step of the process. To evaluate our approach, we use established benchmarks as well as its inclusion in a real-life multiagent-based simulation. Simulated scenarios are based on real customer orders of our industrial partner Hellmann Worldwide Logistics GmbH & Co. KG and are applied on real world infrastructures. The results reveal that efficient optimal decision making in multiagent systems increases the service quality and meets the requirements and challenges in logistics.

Paper Nr: 168
Title:

The Institutional Stance in Agent-based Simulations

Authors:

Giovanni Sileno, Alexander Boer and Tom Van Engers

Abstract: This paper presents a multi-agent framework intended to animate scenarios of compliance and non-compliance in a normative system. With the purpose of describing social human behaviour, we choose to reduce social complexity by creating models of the involved agents starting from stories, and completing them with background theories derived from common-sense and expert knowledge. For this reason, we explore how an institutional perspective can be taken into account in a computational framework. Roles, institutions and rules become components of the agent architecture. The social intelligence of the agent is distributed to several cognitive modules, performing the institutional thinking, whose outcomes are coordinated in the main decision-making cycle. The institutional logic is analyzed from a general simulation perspective, and a concrete possible choice is presented, drawn from fundamental legal concepts. As a concrete result, a preliminary implementation of the framework has been developed with Jason.

Paper Nr: 169
Title:

An Agent-based Model of Autonomous Automated-Guided Vehicles for Internal Transportation in Automated Laboratories

Authors:

Lluís Ribas-Xirgo and Ismael F. Chaile

Abstract: Agent-based modelling enables simulating complex systems and controlling them, as well. In the industrial domain there are plenty of these systems not only because of the size but also because of the need for fault-tolerance and adaptability. Typically, these cases are solved by dividing systems into different dimensions, including the transportation one. In this paper, we take this approach to build a framework to develop and control transportation in applications within the industrial domain, which will be tested on an automated laboratory. The framework is based on a multi-agent simulator that contains the model of the plant with transportation agents having a multi-layered architecture. The lower-level layers correspond to those that would be embedded into physical transportation agents. Therefore, while agents communicate to each other within the simulator environment, communication between upper-level layers and lower-lever layers of each agent is done internally for the simulated parts and externally for the real counterparts. The simulator can be used stand-alone to functionally validate a system or in combination with real agents as a monitoring/controlling tool. Preliminary results prove the viability of the framework as a design tool and show the difficulties to work with physical agents.

Paper Nr: 174
Title:

Evaluation of Emergent Structures in a ”Cognitive” Multi-Agent System based on On-line Building and Learning of a Cognitive Map

Authors:

Abdelhak Chatty, Philippe Gaussier, Ilhem Kallel, Philipe Laroque, Florence Pirard and Adel M. Alimi

Abstract: This paper tries to analyze and evaluate emergent structures in a multi-agent system which is able to resolve the warehouse location problem. These emergent structures allow agents to optimize their planning time and to improve their adaptive behavior in an unknown environment. In our multi-agent system, each agent is based on an on-line building and learning of its own cognitive map. It alters the positive impact of individual behavior in the improvement of the overall performance of the system. We also suggest the evaluation of the emergent structures by comparing the performance of our multi-agent system with a linear programming approach. A series of simulations enables us to discuss and validate our system.

Paper Nr: 178
Title:

Dynamic Agent Prioritisation with Penalties in Distributed Local Search

Authors:

Amina Sambo-Magaji, Inés Arana and Hatem Ahriz

Abstract: Distributed Constraint Satisfaction Problems (DisCSPs) solving techniques solve problems which are distributed over a number of agents.The distribution of the problem is required due to privacy, security or cost issues and, therefore centralised problem solving is inappropriate. Distributed local search is a framework that solves large combinatorial and optimization problems. For large problems it is often faster than distributed systematic search methods. However, local search techniques are unable to detect unsolvability and have the propensity of getting stuck at local optima. Several strategies such as weights on constraints, penalties on values and probability have been used to escape local optima. In this paper, we present an approach for escaping local optima called Dynamic Agent Prioritisation and Penalties (DynAPP) which combines penalties on variable values and dynamic variable prioritisation for the resolution of distributed constraint satisfaction problems. Empirical evaluation with instances of random, meeting scheduling and graph colouring problems have shown that this approach solved more problems in less time at the phase transition when compared with some state of the art algorithms. Further evaluation of the DynAPP approach on iteration-bounded optimisation problems showed that DynAPP is competitive.

Paper Nr: 194
Title:

Real-time Information Querying over Peer-to-Peer Networks using Timestamps

Authors:

Michael Gibson and Wamberto Vasconcelos

Abstract: Most P2P networks are used for file-sharing applications. These forms of applications mainly rely on keyword searching to locate file resources on the peers. Whilst this querying is suitable for many data-intensive applications, it is not suitable for applications where data changes over short periods of time, also known as time-critical applications. We investigate the use of timestamps on a peer’s knowledge about an application to create queries so that other peers may reply with more up-to-date information to keep the peer’s knowledge up-to-date. We propose means to synchronise peers to provide them with a shared, independent clock so that they utilize timestamps. To show that a peer’s knowledge about a time-critical application affects the performance of other peers, we carried out experiments to show information propagation over a P2P network and use various metrics to evaluate our approach.

Paper Nr: 196
Title:

A Self-organizing Multi-Agent System for Combining Method Fragments

Authors:

Noélie Bonjean, Marie-Pierre Gleizes, Christine Maurel and Frédéric Migeon

Abstract: Software systems are becoming more and more complex. A common dilemma faced by software engineers in building complex systems is the lack of method adaptability. However, existing agent-based methodologies and tools are developed for particular system and are not tailored for new problems. This paper proposes an architecture of a new tool based on SME for self-constructing customized method processes. Our approach is based on two pillars: the process fragment and the MAS meta-model. These two elements are both defined and considered under a specific agent-oriented perspective thus creating a peculiar approach. Our work is based on the self-organization of agents, making it especially suited to deal with highly dynamic systems such as the design of an interactive and adaptive software engineering process.

Paper Nr: 201
Title:

Efficient Self Adapting Agent Organizations

Authors:

Kamilia Ahmadi and Vicki H. Allan

Abstract: Self-organizing multi-agent systems provide a suitable paradigm for agents to manage themselves. We demonstrate a robust, decentralized approach for structural adaptation in explicitly modelled problem solving agent organizations. Based on self-organization principles, our method enables the agents to modify their structural relations to achieve a better completion rate of tasks in the environment. Reasoning on adaptation is based only on the agent’s history of interactions. Agents use the history of tasks assigned to their neighbours and completion rate as a measure of evaluation. This evaluation suggests the most suitable agents for reorganization (Meta-Reasoning). Our Selective-Adaptation has four different approaches of Meta-Reasoning, which are 1) Fixed Approach, 2) Need-Based Approach, 3) Performance-Based Approach, and 4) Satisfaction-based Approach along with a Reorganization approach, which needs less data but makes better decisions.

Paper Nr: 227
Title:

Autonomous Timed Movement based on Attractor Dynamics in a Ball Hitting Task

Authors:

Farid Oubbati and Gregor Schöner

Abstract: Timed robotic actions so that they are initiated or terminated just in time can be crucial in many tasks and scenarios in which the robot has to coordinate with other robotic agents or to interact with external entities such as moving objects. The analogy with human movement coordination has motivated an approach in which timed movements are generated from stable periodic solutions of dynamical systems, which are turned on and off in time to initiate and terminate a timed motor act. Here we extend this approach to generate sequences of timed motor actions required to intercept and hit a rolling ball on an inclined plane. The proposed system combines attractor dynamics for the robot’s end-effector heading direction, fixed point attractor for end-effector postural states, limit cycle attractor dynamics for the end-effector speed and competitive neural dynamics to organize the different behaviors and movement phases. The ball interception point and time to contact are predicted based on a Kalman estimate of the ball’s kinematics. The work is implemented on a redundant manipulator CoRA platform and the ball motion is monitored by the manipulator’s vision system.

Paper Nr: 247
Title:

Trusted Community - A Trust-based Multi-Agent Organisation for Open Systems

Authors:

Lukas Klejnowski, Yvonne Bernard, Gerrit Anders, Christian Müller-Schloer and Wolfgang Reif

Abstract: In this paper, the multi-agent organisation Trusted Community is presented. Trusted Communities are formed and joined by self-organised by agents with strong mutual trust relations and the purpose to increase their personal utility. Trusted Communities are maintained by management actions delegated by a designated member called Trusted Community Manager, having the goal to preserve and optimise the composition and stability of this organisation. This organisation provides performance benefits for their members by improving interaction efficiency, information sharing and cooperation between the agents. In the work presented here, Trusted Communities are conceptually defined and the application in an open Desktop Grid System is discussed.

Paper Nr: 251
Title:

Modeling “Info-chemical” Mediated Ecological System by using Multi Agent System

Authors:

Yasuhiro Suzuki and Megumi Sakai

Abstract: We model and simulate an ecological system, where each agent (plants, herbivores and carnivores) communicate with each other by using communication languages of chemical volatiles (info-chemical signals). This info-chemical signals are produced by plants when they are suffered from feeding damage of herbivores and natural enemy (carnivores) of the herbivores are attracted by the signal. In this ecological system, since carnivores learn the signal and trace it, plants try to endure the feeding damage until the population of herbivores become large and enough herbivores can supply for the carnivores, otherwise carnivores are not to be attracted by the signal and try to explore more valuable signal. However, it has reported that, some mutated plants produce chemical signals soon even if there are few or no herbivores and attract carnivores (cry wolf plants). We model the ecological system which cry wolf plants by using the MAS. without geographic space and with geographic space. And we confirm that in the both types of models, in order to escape from cry wolf plants, “honest plants” produce different types of signals so various types of signals emerge. Interestingly, in the system with geographic space, if there is a “colony” of cry wolf plants then signal does not evolve and honest signal and cry wolf signal can coexist.

Paper Nr: 258
Title:

Grateful Agents and Agents that Hold a Grudge - The Role of Affective Behaviors in Sustained Multi-agent Interactions

Authors:

César F. Pimentel

Abstract: Interactions among self-interested agents present classical challenges concerning cooperation and competitiveness. Cooperative behavior may be unappealing if unilateral cooperation represents a loss, and adverse behavior may be difficult to avoid if one agent’s losses imply another agent’s gains. Agents could benefit from mechanisms that promote cooperation and dissuade adverse behaviors. We propose a generic approach, where such mechanisms can emerge from the simulation of affective behaviors that are associated with the human emotions of gratitude and anger. These emotions define implicit contracts about predefined patterns of behavior that agents are capable of following, and recognizing in each other. We use a few examples to illustrate how this approach can help an agent persuade another to cooperate and become an ally, or dissuade it from adopting adverse behavior, as a result of rational decisions.

Posters
Paper Nr: 8
Title:

Right-based Coordination for Environments with Changing Complexity - Results in Traffic Simulation

Authors:

Eduardo Alonso and Peter Kristofersson-Izmajlow

Abstract: In this paper we analyse three different approaches to multi-agent co-ordination, with an accent on the concept of “right-based” agents as opposed to free-rider agents and normative agents. We claim that a balance between unrestricted behaviour and regulatory systems would make collections of agents perform more efficiently, particularly when the complexity and the dynamicity of the environment increases. We present preliminary results on a set of experiments using a traffic simulator.

Paper Nr: 21
Title:

Fault Tolerance through Interaction and Mutual Cooperation in Hierarchical Multi-Agent Systems

Authors:

Rade Stanković and Maja Štula

Abstract: Multi-Agent Systems (MASs) are well suited for development of complex, distributed systems. In its essence MAS is a distributed system that consists of multiple agents working together to solve common problems. Failure handling is an important property of large scale MAS because the failure rate grows with both the number of the hosts, deployed agents and the duration of agent’s task execution. Numerous approaches have been introduced to deal with some aspects of the failure handling. However, absence of centralized control and large number of individual intelligent components makes it difficult to detect and to treat errors. Risk of uncontrollable fault propagation is high and can seriously impact the performance of the system. Although existing research has been extensive, it still needs to attend the MAS failure handling problem in all its aspects, which makes this topic very interesting. We propose a concept of agent interaction that enables any hierarchical MAS to become fault tolerant, regardless of the used agent framework.

Paper Nr: 28
Title:

Evidencing the “Robot Phase Transition” in Human-agent Experimental Financial Markets

Authors:

John Cartlidge and Dave Cliff

Abstract: Johnson, Zhao, Hunsader, Meng, Ravindar, Carran, and Tivnan (2012) recently suggested the existence of a phase transition in the dynamics of financial markets in which there is free interaction between human traders and algorithmic trading systems (‘robots’). Above a particular time-threshold, humans and robots trade with one another; below the threshold all transactions are robot-to-robot. We refer to this abrupt system transition as the ‘robot phase transition’. Here, we conduct controlled experiments where human traders interact with ‘robot’ trading agents in minimal models of electronic financial markets to see if correlates of the two regimes suggested by Johnson et al. (2012) occur in such laboratory conditions. Our results indicate that when trading robots act on a super-human timescale, the market starts to fragment, with statistically lower human-robot interactions than we would expect from a fully mixed market. We tentatively conclude that this is the first empirical evidence for the robot phase transition occurring under controlled experimental conditions.

Paper Nr: 31
Title:

Filtering Relevant Facebook Status Updates for Users of Mobile Devices

Authors:

Stephan Baumann, Rafael Schirru and Joachim Folz

Abstract: In recent years, social networking sites such as Twitter, Facebook, and Google+ have become popular. Many people are already used to accessing their individual news feeds ubiquitously also on mobile devices. However the number of status updates in these feeds is usually high thus making the identification of relevant updates a tedious task. In this paper we present an approach to identify the relevant status updates in a user’s Facebook news feed. The algorithm combines simple features based on the interactions with status updates together with more sophisticated metrics from the field of Social Network Analysis as input for a Support Vector Machine. Optionally the feature space can be extended by a topic model in order to improve the classification accuracy. A first evaluation conducted as live user experiment suggests that the approach can lead to satisfying results for a large number of users.

Paper Nr: 47
Title:

Group Formation and Knowledge Sharing in Pedestrian Egress Simulation

Authors:

Kyle D. Feuz and Vicki H. Allan

Abstract: Pedestrian simulation has been a topic of research for several decades, especially in regards to pedestrian egress. Only recently, though, have researchers begun to consider the effects that groups have upon pedestrian egress. Both empirical studies and simulation models predict a decrease in pedestrian speeds when pedestrians travel in groups. In this study, we show that this decrease in speed does not necessarily correspond to an increase in egress time as additional factors such as the amount of knowledge gained through the formation of groups must be considered. The sharing of route costs helps pedestrians maintain proximity to each other and under certain circumstances, pedestrian egress times are actually improved by the formation of groups. We also show that the inclusion of communication costs, sharing knowledge, and group decision-making all have a strong impact on predicted egress times.

Paper Nr: 58
Title:

Performance Evaluation for Autonomous Mobile Robots

Authors:

David Trejo, Nelson Biedma, Daniela López De Luise, Lucas Rancez, Gabriel Barrera and Leonardo Isoba

Abstract: The aim of this paper is to implement metrics and to define indicators to provide a unified criteria for evaluation method of the performance of autonomous mobile robots using different control algorithms. There is a first description of the mobile robot problem and the importance of a standardized process for evaluation of robot performances. There is a comparison between a simple navigation controller and an intelligent prototype based on consciousness. The architecture main features are also outlined. Test cases with and without conscious controller show that the latter performs an optimized source-to-target path.

Paper Nr: 78
Title:

Autonomous Aquatic Agents

Authors:

A. Calce, P. Mojiri Forooshani, A. Speers, K. Watters, T. Young and M. Jenkin

Abstract: Constructing a collection of autonomous agents requires the development of appropriate experimental hardware platforms. Here we describe the process of re-purposing inexpensive radio-controlled (RC) electric motorboat as autonomous surface craft. Standard electronics components are used to interface with the RC boat electronics, and the vessels are augmented with GPS, vision, and a tilt-compensated compass to provide the necessary onboard sensing capabilities to enable point-to-point and target-based control of the vehicle. A ROS-based control and sensing infrastructure is used to operate the vehicles on-board while 802.11n communication provides communication off-board. Vessels have been operated successfully in both the pool and ocean environment.

Paper Nr: 84
Title:

Information System for Autonomous Mobile Robot Interaction - Position Paper

Authors:

Nadina Battagliotti, Daniela López De Luise and Jin Sung Park

Abstract: This paper presents what the authors consider to be a new distributed memory architecture for self-organizing distributed systems. The work in progress presented here focuses on the representation of a distributed memory for multi-robot systems. It states the main characteristics of a shared environment and provides suitable interfaces to allow autonomous mobile robots to query, publish and exchange relevant information without requiring a central data storage. The memory is based on a combination of fuzzy systems, distributed Self-Organizing Maps (SOM), a new specific design information handshake between robots and a new model for approach for world-map global administration.

Paper Nr: 92
Title:

HTN Planning for Pick-and-Place Manipulation

Authors:

Hisashi Hayashi, Hideki Ogawa and Nobuto Matsuhira

Abstract: We introduce new heuristics of HTN (Hierarchical Task Network) planning for mobile robots with two arms/hands that pick and place objects among movable obstacles. Based on our new heuristics, the robot moves obstacles if necessary, picks and places the target objects without collisions. The robot chooses the (right or left) hand to use for each manipulation in order to avoid collisions and reduce the number of obstacle movements. In most of the previous approaches that combine task planning and motion planning, collisions between an arm and obstacles are checked only by the lower-level geometric motion planner. Therefore, the high-level general-purpose task planner often produces a plan that is not executable by the lower-level modules. On the other hand, in our new heuristics, the task planner roughly checks collisions, and produces executable plans.

Paper Nr: 122
Title:

Controlling Complex Systems Dynamics without Prior Model

Authors:

Jérémy Boes, François Gatto, Pierre Glize and Frédéric Migeon

Abstract: Controlling complex systems imposes to deal with high dynamics, non-linearity and multiple interdependencies. To handle these difficulties we can either build analytic models of the process to control, or enable the controller to learn how the process behaves. Adaptive Multi-Agent Systems (AMAS) are able to learn and adapt themselves to their environment thanks to the cooperative self-organization of their agents. A change in the organization of the agents results in a change of the emergent function. Thus we assume that AMAS are a good alternative for complex systems control, reuniting learning, adaptivity, robustness and genericity. The problem of control leads to a specific architecture presented in this paper.

Paper Nr: 137
Title:

A MDA Methodology to Support Multi-Agent System Development

Authors:

Carlos Eduardo Pantoja and Ricardo Choren

Abstract: Currently, Multi-Agent Systems (MAS) technologies and platforms are either emerging or changing constantly, which implies a high effort in developing software products. This scenario may generate several problems related to software modeling, design, coding, integration and interoperability. The Object Management Group (OMG) proposes the Model Driven Architecture (MDA), which defines an architecture based on platform independent models (PIM) and platform specific models (PSM). In this work we present a model-driven methodology, based on MDA, for the development of MAS. The methodology proposes the use of agent-based PIM and PSM, and defines mapping rules between models. The paper also presents a simple case study to illustrate the proposal.

Paper Nr: 170
Title:

An Interface Agent for the Management of COTS-based User Interfaces

Authors:

Jose F. Sobrino, Javier Criado, Jesus Vallecillos, Nicolas Padilla and Luis Iribarne

Abstract: The great development of the knowledge society on the Internet requires that Web information systems are adapted at runtime to user groups with common interests. Interface agents help us to observe and learn from user preferences making interfaces adaptable to user working habits. We propose an interface agent which works on Web interface based on COTS components, adapting the interface to the user needs or preferences. Our agent runs two main behaviors: observation behavior which analyses the user interaction on the interface and a second behavior which runs the adaptation actions to adapt the user interface at runtime.

Paper Nr: 175
Title:

Simulation of University Education Process

Authors:

Jiří Jelínek

Abstract: The article deals with one possible usage of agent based simulation. This method is advantageous for tasks that can be modeled through a set of basic building elements - agents and their interaction. This contribution shows an application of this approach for the simulation of university education. Conceptual model of university education based on AB modeling of the students was developed for this purpose. The aim was to develop a tool which could help to improve the quality of higher education processes through better knowledge about them obtained from their simulations.

Paper Nr: 181
Title:

A Multi-robot System for Patrolling Task via Stochastic Fictitious Play

Authors:

Erik Hernández, Antonio Barrientos, Jaime del Cerro and Claudio Rossi

Abstract: A great deal of work has been done in recent years on the multi-robot patrolling problem. In such problem a team of robots is engaged to supervise an infrastructure. Commonly, the patrolling tasks are performed with the objective of visiting a set of points of interest. This problem has been solved in the literature by developing deterministic and centralized solutions, which perform better than decentralized and non-deterministic approaches in almost all cases. However, deterministic methods are not suitable for security purpose due to their predictability. This work provides a new decentralized and non-deterministic approach based on the model of Game Theory called Stochastic Fictitious Play (SFP) to perform security tasks at critical facilities. Moreover, a detailed study aims at providing additional insight of this learning model into the multi-robot patrolling context is presented. Finally, the approach developed in this work is analyzed and compared with other methods proposed in the literature by utilizing a patrolling simulator.

Paper Nr: 185
Title:

Formalizing SIMBA RTMAS Models using Real-time Maude

Authors:

Toufik Marir, Farid Mokhati and Hassina Seridi-Bouchelaghem

Abstract: Multi-agent systems paradigm is the most appropriate one to developing complex systems. Consequently, it is used to develop real-time intelligent systems which are one of complex systems categories. In the literature, several works have been proposed for formalizing many aspects of multi-agent systems. However, the application of formal methods upon real-time multi–agent systems (RTMAS) (where the time is the primordial aspect) stills in the immaturity stage. In this paper, we present the formalization of SIMBA real-time multi-agent systems using Real-Time Maude language as a main stone for formal development of based-SIMBA systems.

Paper Nr: 186
Title:

Strategic Reorganization in Geographically Dispersed Multi-Agent Systems - A Case Study using Cow Herding Scenario

Authors:

Maryamossadat N. Mahani and Arvin Agah

Abstract: In this work we address the problem of coordinating a group of agents to execute a set of geographically dispersed tasks in an uncertain environment. In such domain, where there are distinct task types to be performed, we argue that there is inherent leverage to enforcing reorganization. We consider a distributed problem solving application in which agents must coordinate and cooperate in groups to complete different task types. While each group can benefit from a certain structure of agents depending on the task in hand, the general goal is to increase the utility gained over a given time horizon. Our approach to this problem is based on keeping an advanced knowledge of organizational structures appropriate for certain task types considering the geographical dispersion. We perform experiments and report on the results.

Paper Nr: 189
Title:

Self-adaptive Aided Decision-making - Application to Maritime Surveillance

Authors:

Nicolas Brax, Eric Andonoff, Marie-Pierre Gleizes and Pierre Glize

Abstract: Information required for decision-making in complex applications, such as flood forecast or maritime surveillance, can be represented using a mathematical function. However, due to the complexity of the considered applications and their dynamics, the parameters involved in the mathematical function can be hard to value a priori. This paper presents a Multi-Agent System, called PaMAS (Parameter Multi-Agent System) that is able to learn such parameters values on the fly, autonomously, cooperatively and by self-adaptation. It also illustrates the application of PaMAS in the context of the maritime surveillance European project I2C. It finally provides an evaluation of the PaMAS learning.

Paper Nr: 212
Title:

Handling Exceptions in Multi Agent Systems using Learning Agents

Authors:

Mounira Bouzahzah and Ramdane Maamri

Abstract: In this paper we address the problem of exception handling in multi agent systems and we propose an approach based learning agents to provide fault tolerance in multi agent systems. In reality, agents systems are used as a perfect solution to design recent applications that are characterized by being decentralized and dispersed. These systems are subject of errors that occur during execution and that may cause system’s failure; Exceptions are the main cause of the system’s errors. Researchers in fault tolerance field use handling exceptions technique to provide error-prone systems. Through this work, we propose an approach for handling exception using learning agents; this approach assures the most efficient handler for each exception mainly in case of the existence of many handlers and allows the adaptation of decision according to the environment changes. The learning agent is given the capacity to learn about new exceptions from the extern.

Paper Nr: 230
Title:

Trekking Navigation System using Opportunistic Communication

Authors:

Yasuhiko Kitamura, Shunsuke Nosaka, Hirofumi Kishino and Yui Okuda

Abstract: Mountain climbing or trekking becomes popular in Japan recently. Unfortunately the more climbers go to mountains, the more get lost. Handy GPS’s are a well-known tool to navigate climbers in the mountain area by displaying their location on a digital map. They are useful to know the current location, but not suitable to call an emergency help. Cell phones are useful to call a help, but they work only in the city area accessible to mobile phone networks. They seldom get access to them in the mountain area. This paper proposes a new trekking navigation system, which consists of mobile terminals and a server, that works even in the poor communication environment where the Internet access is often disrupted such as in the mountain area. The terminals can navigate climbers even when the Internet access is unavailable. When they can get access to the server, they send the walking trajectory of the climbers, so the rescue party can locate the climbers in need. They can also exchange the walking trajectory with each other by utilizing opportunistic communication and carry the information until they reach an area accessible to the server. This paper shows a prototype of trekking navigation system under development and how the opportunistic communication improves the location estimation of climbers.

Paper Nr: 232
Title:

Auction Model of P2P Interaction in Multi-Agent Software

Authors:

Anton Ivaschenko and Andrey Lednev

Abstract: The paper describes one of the possible models of interaction management of active software agents in P2P network of the enterprise information space. The matrix form of enterprise management is being projected on the P2P network of interacting software components. It is proposed to study the problem of resource allocation using Auction model enhanced by the opportunity of its management by changing dynamical characteristics. The approach shows that the introduction of delays and accelerations allows to increase the efficiency of solving scheduling problems.

Paper Nr: 242
Title:

Comparison and Multi-GPU based Implementation of a Collision Detection Method with Z-buffer in Cyber Space

Authors:

Tomoaki Iida, Hidemi Yamachi, Yasushi Kambayashi and Yasuhiro Tsujimura

Abstract: We propose a new technique to detect objects and collisions of geometric objects in cyber space. This technique uses depth values of the Z-buffer in rendering scene, and contributes to construct an intelligent interface for mobile software agent that uses augment reality. We use the orthographic projection for collision detection. Our method uses two depth value sets. One is obtained through rendering the cyber space from the sensor object toward a target point. This set does not have the depth values of the sensor object. Another one is obtained through rendering only the sensor object in the reverse direction. From these two depth value sets, we obtain the distance between the sensor object and others for each pixel. This technique requires only one or two rendering processes and it is independent from the complexity of object's shape, deformation or motion. In this paper we evaluate the efficiency in comparison with one of the most popular collision detection method. GPU based methods have advantage for utilizing multi-GPU system. In order to take the advantage, we have implemented our method for multi-GPU system and evaluate the performance of collision detection.

Paper Nr: 243
Title:

Design of an Intelligent Interface for the Software Mobile Agents using Augmented Reality

Authors:

Kazuto Kurane, Munehiro Takimoto and Yasushi Kambayashi

Abstract: In this paper we propose an intelligent interface for the mobile software agent systems that we have developed. The interface has two roles. One is to visualize the mobile software agents using augmented reality (AR) and the other is to give human users a means to control the mobile software agents by gesture using a motion capture camera. Through the interface we human beings can intuitively grasp the activities of the mobile agents, i.e. through augmented reality. In order to provide proactive inputs from user, we utilize a Kinect motion capture camera to recognize the human users’ will. The Kinect motion capture camera is mounted under the ceiling of the room, and they monitor the user as well as robots. When the user points at a certain AR image, and then points at another robot through gesture, the monitoring software recognizes the will of the user and conveys instructions to the mobile agent based on the information from the Kinect. The mobile agent that is represented by the image moves to the robot that was pointed. This paper reports the development of the intelligent user interface described above.

Paper Nr: 256
Title:

A Method of Conceptual Modelling for Realistic Training Scenarios

Authors:

Inna Shvartsman, Kuldar Taveter and Merik Meriste

Abstract: Training is crucial for improvement of the capabilities of both military and non-military personnel. In this paper, we argue for the need of conceptual modelling for the creation of training scenarios. This research proposes a particular method for developing training scenarios for complex domains based on agent–oriented modelling. The advantage of agent-oriented modelling is that it enables to describe a problem domain from three balanced and interrelated aspects – interaction, information, and behaviour, and at three abstraction layers: analysis, design, and simulation. Thus, we can obtain agent-directed simulations for elaborating selected aspects of emergent behaviour to support development of practical training scenarios in a partially known environment.