Advances in Multi-Agent Reinforcement Learning Intelligence

Dear Colleagues,

Multi-Agent Reinforcement Learning (MARL) has emerged as a powerful paradigm to enable intelligent agents to learn and adapt independently in dynamic, multi-agent systems. With applications ranging from autonomous robotics, intelligent transport, finance, distributed control systems, and multi-robot coordination, MARL has demonstrated its potential to tackle difficult decision-making problems where multiple agents must coordinate, compete, or negotiate in uncertain and partially observable worlds.

Despite impressive progress, some of the most significant challenges still pervade MARL research, including:

• Coordination and Communication: Designing efficient mechanisms through which agents can convey beneficial information without sacrificing scalability and robustness in large systems.
• Scalability and Generalization: Scaling MARL algorithms to an increasing number of agents, transferring knowledge across different tasks, and generalizing appropriately to new environments.
• Exploration and Sample Efficiency: Reducing computational expenses and improving the efficiency of learning by leveraging advanced exploration techniques and reward shaping.
• Trust, Safety, and Explainability: Improving interpretability, fairness, and trustworthiness of MARL applications, particularly in high-stakes domains such as healthcare, finance, and autonomous vehicles.

The rapid evolution of deep reinforcement learning, coupled with advances in multi-agent coordination mechanisms, has opened up new research opportunities. New avenues in graph-based MARL, attention-based communication mechanisms, and curriculum learning for agent training are paving the way for more effective and scalable solutions. Furthermore, the application of MARL to actual decentralized networks, swarm robotics, and smart infrastructures continues to present research challenges regarding efficiency, adaptability, and security.

This Special Issue invites researchers to submit original research articles, case studies, and detailed reviews that encompass both the theoretical innovation and applied aspects of MARL.

Submissions on novel algorithmic developments, new applications, and cross-disciplinary breakthroughs that extend the boundaries of intelligent multi-agent systems are welcomed.

Dr. Sergio Spanò
Dr. Luca Di Nunzio
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Electronics is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• multi-agent reinforcement learning (MARL)
• embedded intelligence
• distributed control systems
• real-time processing
• autonomous systems