TIM-MARL: Information Sharing for Multi-Agent Reinforcement Learning in Smart Environments
Abstract
Information sharing among agents to jointly solve problems is challenging for multi-agent reinforcement learning algorithms (MARL) in smart environments. In this paper, we present a novel information sharing approach for MARL, which introduces a Team Information Matrix (TIM) that integrates scenario-independent spatial and environmental information combined with the agent's local observations, augmenting both individual agent's performance and global awareness during the MARL learning. To evaluate this approach, we conducted experiments on three multi-agent scenarios of varying difficulty levels implemented in Unity ML-Agents Toolkit. Experimental results show that the agents utilizing our TIM-Shared variation outperformed those using decentralized MARL and achieved comparable performance to agents employing centralized MARL.
Department(s)
Computer Science
Document Type
Conference Proceeding
DOI
10.1109/CCNC51664.2024.10454813
Keywords
Deep reinforcement learning, hierarchical information sharing, multi-agent system, Unity ML-Agent Toolkit
Publication Date
1-1-2024
Recommended Citation
Siddiqua, Ayesha; Liu, Siming; Iqbal, Razib; Irfan, Fahim Ahmed; Ross, Logan; and Zweerink, Brian, "TIM-MARL: Information Sharing for Multi-Agent Reinforcement Learning in Smart Environments" (2024). Faculty Scholarship. 484.
https://bearworks.missouristate.edu/articles00/484
Journal Title
Proceedings IEEE Consumer Communications and Networking Conference Ccnc
