Multi-Agent-Based Simulation XXI: 21st International Workshop, MABS 2020, Auckland, New Zealand, May 10, 2020, Revised Selected Papers (Lecture Notes in Computer Science, 12316)

This book constitutes the thoroughly refereed post-conference proceedings of the 20th International Workshop on Multi-Agent-Based Simulation, MABS 2020, held in Auckland, New Zealand, in May 2020 collocated with 19th International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS 2020). Due to COVID-19 the workshop has been held online. The 9 revised full papers included in this volume were carefully selected from 11 submissions. The workshop focused on finding efficient solutions to model complex social systems, in such areas as economics, management, organizational and social sciences in general and much more. Preface Organization Contents Adaptivity in Distributed Agent-Based Simulation: A Generic Load-Balancing Approach 1 Introduction 2 Adaptivity in Agent-Based Simulation 3 Distributed Simulation Architecture: Acsim 4 MAPE-K as a Generic Framework for Adaptivity 5 Motivating Examples 5.1 Adaptive Local Optimization of Compute Cost - A Micro-traffic Example 5.2 Adaptive Local Optimization of Communication Cost - A Cellular Automata Example 6 Conclusion References Trajectory Modelling in Shared Spaces: Expert-Based vs. Deep Learning Approach? 1 Introduction 2 Methodology 2.1 Problem Formulation 2.2 Game-Theoretic Social Force Model 2.3 LSTM with DBSCAN 3 Data Sets and Evaluation Metrics 3.1 Data Sets 3.2 Evaluation Metrics 4 Experimental Results 4.1 Quantitative Results for Individual Models 4.2 Qualitative Results for Individual Models 4.3 Pros and Cons of GSFM and LSTM-DBSCAN 5 Conclusion and Future Work References Towards Agent-Based Traffic Simulation Using Live Data from Sensors for Smart Cities 1 Introduction 2 Collecting Traffic Counts 3 The Liverpool Smart Pedestrians Project 4 An Edge Computing Device for Traffic Monitoring 5 Using Live Data in Traffic Simulation 6 Preliminary Results 7 Conclusion and Future Work References Design and Evaluations of Multi-agent Simulation Model for Electric Power Sharing Among Households 1 Introduction 2 Related Works 3 Assumption of Electric Power Sharing 3.1 Outline of Electric Power Sharing Service 3.2 Assumptions 4 Design of MAS Model 4.1 Household Agent 4.2 Electricity Retailer Agent 5 Basic Evaluation of Electric Power Sharing MAS Model 5.1 Evaluation Policy 5.2 Evaluation Indicator 5.3 Input Data 5.4 Conditions for Basic Evaluation 5.5 Evaluation Results 5.6 Analysis 6 Conclusion References Active Screening on Recurrent Diseases Contact Networks with Uncertainty: A Reinforcement Learning Approach 1 Introduction 2 Problem Statement 3 Challenges 4 Partially Observable States 5 Combinatorial Maximization Problem 5.1 Incremental Selection 5.2 One-Shot Selection 6 High-Dimensional State-Action Space 6.1 One-Hot Encoding 6.2 Nodewise Encoding 6.3 Graph Convolutional Neural Networks 7 Experiments 7.1 Experiment Setup 7.2 Experiment Result 8 Conclusion 9 Appendix 9.1 Belief Update References Impact of Meta-roles on the Evolution of Organisational Institutions 1 Introduction 2 An Overview of the Extended BDI Architecture 3 Meta-roles and Role Dynamics 4 Simulation, Algorithms, and Parameters 4.1 Assumptions 4.2 Societies 4.3 Algorithms 4.4 Parameters 5 Results 5.1 Permissions for Private Trade 5.2 Fired Violators (Monitoring Strength) 6 Discussion and Concluding Remarks References Optimization of Large-Scale Agent-Based Simulations Through Automated Abstraction and Simplification 1 Motivation 2 Related Work 3 Simplification Types 3.1 Subsampling 3.2 Simplifying Agents' Decision Process 3.3 Selectively Replacing Groups of Agents with One Agent 3.4 Simplifying Communication Between Agents 4 Experiments 4.1 Simplifications Support in FARM 4.2 GitHub and Twitter Simulations 4.3 Experiment Setup 5 Results 6 Semi-automated Search for Effective Simplifications 7 Conclusions and Future Work References Improved Travel Demand Modeling with Synthetic Populations 1 Introduction 2 State of the Art in Mobility Modeling 3 The Synthetic Population Approach 4 Comparison of Results 4.1 Trip Counts Comparison 4.2 Departure Time Comparison 4.3 Spatial Distribution Comparison 5 Conclusion References Author Index