Andrei Olaru – författare

Development of Multi-Agent Systems Infrastructure: A Practical Approach explores the creation of modular frameworks to support the deployment of real-world software applications utilizing multi-agent systems (MAS). Drawing from the author’s hands-on experience with the FLASH-MAS framework-a Fast Lightweight Agent Shell-the book delves into both theoretical models and practical solutions for MAS implementation. It addresses the complexities of deploying autonomous agents across diverse fields such as manufacturing, robotics, health care, and supply chain management, highlighting the shared challenges developers face when managing distributed, networked, or large-scale agent interactions. The book is organized into three main sections, covering models and languages for MAS, the deployment and interaction between system entities, and practical guidance for implementing robust MAS frameworks. Emphasizing modularity, the author presents adaptable tools and solutions that can be independently utilized for system development and maintenance. Practical issues such as entity lifecycle management, environmental interactions, and system robustness are thoroughly examined, making this resource valuable for both new and experienced MAS developers.Presents models and languages for multi-agent systems (MAS) developmentIncludes three key sections covering: modeling and languages for multi-agent systems; entity deployment and interaction; and practical considerations for implementing contemporary MAS frameworksProvides a hands-on method for developing working MAS solutions across a variety of applicationsShowcases solutions in a structured, modular format so they can be replicated and used independently from one another

Distributed AI in the Modern World: Technical and Social Aspects of Interacting Intelligent Agents presents state-of-the-art insights into the various forms of distribution of artificial intelligence, with practical application instances. Sections provide readers with practical solutions at an architectural level, with solutions presented on the distribution of the learning process and the utilization of machine learning models in a distributed system, tools that enable the distribution and interaction of artificial learning entities, how multi-agent systems and machine learning can be combined, the physical embodiment of intelligent agents, and the interaction of intelligent computing units bound to physical space.Following sections emphasize the challenges that are common to all scenarios and solutions that apply in a wider range of cases. This book does not analyze the internal workings of machine learning models (for instance, in the case of multi-agent reinforcement learning), but instead provides readers with an overview of the challenges brought by the need of artificially intelligent entities to interact with other entities and with their environments, along with practical solutions at an architectural level.Presents leading-edge insights on the diverse ways artificial intelligence is distributed, along with real-world application examplesOffers readers a summary of the challenges posed by the necessity for artificially intelligent entities to engage with other entities and their surroundingsProposes practical architectural-level solutionsExplores communication in local networks, on the web, and within physical environmentsExamines machine learning distributed across a network into software agents using machine learning models to make decisions, and into agents working on the web, in socio-technical networks, and in AI embodied in robots

This two-volume set LNCS16258-16259 constitutes the refereed proceedings of the 22nd European Conference on Multi-Agent Systems, EUMAS 2025, held in Bucharest, Romania, during September 3–5, 2025.The 33 full papers, 9 short papers, 3 demonstrators and 1 Agent Toolkits community report included in this book were carefully reviewed and selected from 60 submissions. They were organized into the following topical sections:Part I: Engineering Multi-Agent Systems; Agent-Based Modeling and Simulation; Human-Agent Interaction; and Agent Toolkits Community Report.Part II: Formal approaches, Networks and Learning; Short Papers; and Demonstrators.