Design and Applications of Intelligent Agents: Third Pacific Rim International Workshop on Multi-Agents, PRIMA 2000 Melbourne, Australia, August ... (Lecture Notes in Computer Science, 1881)

PRIMA 2000 was the third in the series of Paci c Rim International Workshops on Multi-Agents. It was held on August 28-29, 2000, in Melbourne, Australia in conjunction with the Paci c Rim International Conference on Arti cial Intel- gence 2000. PRIMA is the main forum for the agent or multi-agent researchers in paci c rim countries to exchange and discuss their research results. This volume contains selected papers from PRIMA 2000. It covers theory, design, and applications of intelligent agents. The speci c aspects include co- dination, negotiation, learning, architecture, speci cation, allocation, and app- cation of intelligent agents. All papers are of high quality because each of them was reviewed and recommended by at least two international renowned program committee members. Many people contributed to this volume. We would like to thank all the a- hors who submitted papers to the workshop. Many thanks also to the members of the program committee who diligently reviewed all the papers. Finally, we thank the editorial sta of Springer-Verlag for publishing this volume in the Lecture Notes in Arti cial Intelligence series. Design and Applications of Intelligent Agents Preface Table of Contents Coordinating Distributed Decision Making Using Reusable Interaction Specifications Agent Coordination Conclusion Acknowledgements. References Coordination Strategy Representation Coordination Context Coordination Strategy Construct Role Construct Strategy Selection Strategy Characteristics Analysis Selection Mechanism Introduction Dynamic Properties of Multiagents Based on a Mechanism of Loose Coalition Introduction Agent Model Based on Subjective Information Subjective Information Behavior of Agents Based on Subjective Information Loose Coalition for Multiagent Systems Definition of Loose Coalition Behavior of Multiagent Systems Based on Loose Coalition Experiment on Dynamics of Multiagent System Based on Loose Coalition Properties and Functions of Agents System Configuration Experimental Results Evaluation Conclusion References Agent Negotiation under Uncertainty and Risk 1 Introduction 2 Assumptions under Trusted Third Party Mediated Uncertain Games 3 The Communication Mechanisms and Negotiation Protocol 3.1 The General Negotiation Protocol 3.2 Negotiation Protocol - Make an Offer 3.3 Negotiation Protocol -- Accept/Reject/Counter Offer 4 A Scenario of Negotiation in an Uncertain Game Risk Preference Agents 5 Discussion 5.1 The Effects of Different Risk Preferences on the Negotiation Results 5.2 The Properties of Negotiation Protocol 6 Conclusion Acknowledgment References Argumentation as a Social Computing Paradigm Introduction An Argumentation Framework Knowledge Representation Argument Multi-paradigm Logic Argument-Based Reasoning Conflict-Resolving Reasoning Dialectical Reasoning Cooperative Reasoning Integrating Three Reasoning Methods into the Argumentation Protocol A Variety of Applications Applications of Conflict-Resolving Reasoning Applications of Dialectical Reasoning Applications of Cooperative Reasoning Concluding Remarks Acknowledgments. References Rationality Assumptions and Optimality of Co-learning Introduction and Background Extensive Games Reinforcement Learning Levels of Rationality Assumptions ``Double-Estimation" Reinforcement Learning Some Experiments The Left/Right (LR) Game The Repeated Prisoner's Dilemma (RPD) Game The Poker (P) Game Discussions References Experiences in the Development of an Agent Architecture Introduction Background The Initial Architecture Lessons Learned Control, Monitoring, and Organisation Tools Matching Message Content to Internal Storage Supporting Agent Conversations Recommendations for Further Work Conclusions References Building a Network Community Support System on the Multi-agent Platform Shine Introduction Community Organizer textsf {Shine}: A Multi-agent Platform for Network Community Support Systems Multi-agents for Network Community Applications Internal Structure of a textsf {Shine} Agent Comparisons with Other Multi-agent Platforms Community Organizer on textsf {Shine} Conclusions References A Secure Route Structure for Information Gathering Agent Introduction Protecting an Agent's Route A Simple Route Structure A Secure and Robust Route Structure Cooperative Protection of Agent's Route - Basic Model Protocol to Obtain the Location of the Alternative Host Security Issues An Extension to the Basic Model Application Scenario Electronic Commerce Intrusion Detection Future Works Conclusion References An Approach to Specifying Coordinated Agent Behaviour 1 Introduction 2 The SimpleTeam Programming Framework 2.1 The Team Construct 2.2 The Role Construct 2.3 Coordinating Sub-team Activity 3 Example 4 Discussion Automated Team Formation Failure Detection and Recovery Specification of Agents' Influences 5 Conclusion References Specifying Agent Behaviour with Use Cases Introduction Use Case Analysis for Requirements Specification Use Cases for Non Agent Applications Use Cases for Agent Applications Methodology for Specifying Agent Systems Identify Agents and Actors Identify Use Cases for Agents and Actors Document Each Use Case Case Study Domain Description Identifying Actors Identifying Use Cases Documenting Use Cases Discussion and Future Work Note about the BDI Agent Model Human-in-the-loop Variant More Complex Systems Conclusions References Market-Based Network Resource Allocation with Non-tâtonnement Process Introduction An Approach Based on Non-t{accent 94 a}tonnement Process Formalization of Non-t{accent 94 a}tonnement Process Calculation Based on Non-t{accent 94 a}tonnement Process Comparison of T{accent 94 a}tonnement Process and Non-t{accent 94 a}tonnement Process Experiments Configuration Results Discussion Conclusion References Autonomous Cooperative Factory Control 1 Introduction 1.1 Technology to Enable Flexibility 1.2 The Autonomous Cooperative System 2 Architecture of an Autonomous Cooperative System (ACS) 2.1 Physical Architecture 2.2 Product ACU 2.3 Equipment ACU 2.4 ACU Interaction 3 Application of ACS to a Factory 3.1 Rod Mill 3.2 Steel Rod Production 3.3 Compensation through Cooperation 4 Simulation Results 4.1 Simulation Testing 4.2 Equipment Wear 4.3 Equipment Failure 5 Field Trial Results 6 Conclusion 7 References Delegation of Responsibility in an Agent-Based Process Management System Introduction Process Management The Multi-agent System Delegation Strategy Choosing the ‘Best’ Individual Parameters Estimates Conclusion References Author Index

This volume constitutes the refereed proceedings of the Third Pacific Rim International Workshop on Multi-Agents, PRIMA 2000, held in Melbourne, Australia in August 2000 in conjunction with PRICAI 2000. The 13 revised full papers presented were carefully reviewed and selected for inclusion in the book. The papers are organized in topical sections on coordination, negotiation, and learning; architecture, specificiation, and allocation; and applications in control and process management.