This book constitutes the refereed proceedings of the 23rd International Conference on Practical applications of Agents and Multi-Agent Systems, PAAMS 2025, held in Lille, France, during June 25 27, 2025. The 27 full papers and 9 short papers included in this book were carefully reviewed and selected from 57 submissions. The PAAMS 2025 proceedings focus on application and validation of agent-based models, methods, and technologies in a number of key application areas, including computational social science, digital twins, simulating complex systems, agents for social good, advanced models for learning, agent-based programming, distributed data analysis, automatic planning, decision-making, social interactions, formal and theoretic models, self-adaptation, mobile edge computing, swarms, and task allocation.

Inhalt

.- Main Track.
.- Transformer Based Models for Offline Multi-Agent Reinforcement Learning.
.- Theory and Practice of Natural Strategy Checking in Concurrent Games with Imperfect Information.
.- Optimizing road intersections using phase scheduling.
.- SysTemp: A Multi-Agent System for Template-Based Generation of SysML v2.
.- Evaluating Infrastructure-Less Cooperative Parking Guidance via Agent-Based Simulation.
.- Multi-Agent Reinforcement Learning with Automated Negotiation for Serial Supply Chain Management.
.- Royal FLush: a MAS-based platform for Decentralized Federated Learning based on SPADE agents.
.- Prioritized Path Planning for Multi-Agent Pickup and Delivery with Multi-Item Packing Problem.
.- Distribution Model: Separation of Concerns to Facilitate the Distribution of Agent-Based Models.
.- Towards an LLM-powered Social Digital Twinning Platform.
.- Knowledge Graph-Enhanced Multi-Agent Infrastructure for coupling physical and digital robotic environments.
.- Applying an Innovative TCN and Multi-Tasks SVM for Early Sepsis Prediction.
.- Towards understanding Complex Adaptive Systems: An illustration on urban logistics.
.- Efficient Assignment of Immediate Tasks Using Deep Reinforcement Learning in Multi-Agent Pickup and Delivery.
.- Towards a replicable synthetic population and agent-based transport model for Bavaria.
.- LDAS: Proposal for semi-optimal Storage Locations using LDA and SA
.- Influence of State Representation on Algorithmic Collusion under Deep Learning.
.- MASPY: A Python-Based Framework for Developing BDI Multi-Agent Systems.
.- Is Generative Artificial Intelligence Ready for Computational Social Science?.
.- Learning Controllability Using Destination Channels with Reward Navigation in Multi-Agent Systems.
.- A migration control approach to improve security for Open MAS in IoT networks.
.- Drone Swarms for Multi-Perspective Monitoring of Large Mammals in their Natural Habitats: Deployment and Field Trials.
.- An Identity Based Agent Model for AI Value Alignment.
.- Rasa4JaCa - A Communication Interface between Multi-agent Systems and Opensource Chatbot Technologies.
.- SPADE as a MAS framework for the development of Digital Twins.
.- TimeCIEL : Contextual Interactive Ensemble Learning for Time Series Classification.
.- Marketing Strategies and Context Effects.

.- DEMONSTRATIONS TRACK.
.- Why is intersection control such a difficult task? A demonstration.
.- Situated Agents in Action: Extending VEsNA with Spatial and Grabbable Artifacts.
.- On the Usability and Extensibility of VITAMIN.
.- Analysis, Implementation and Demonstration of the Nim Game Mathematical Winning Strategy.
.- Machine learning for decision support and automation in games: agent city navigation.
.- Planner-Guided Robot Swarm Demonstration with Natural Language Control.
.- Demonstration: WellProdSim Social Simulator Terrain Model.
.- From Console to Dashboard: A Demonstration of Real-Time Visualization for Social Simulation with WellProdSim.
.- CyberAIMs: Using Agent-based Modeling & Simulation to address Strategies in Cyberspace.