Optimization for Reinforcement Learning: From Single Agent to Cooperative Agents

Abstract

This article reviews recent advances in multi-agent reinforcement learning algorithms for large-scale control systems and communication networks, which learn to communicate and cooperate. We provide an overview of this emerging field, with an emphasis on the decentralized setting under different coordination protocols. We highlight the evolution of reinforcement learning algorithms from single-agent to multi-agent systems, from a distributed optimization perspective, and conclude with future directions and challenges, in the hope to catalyze the growing synergy among distributed optimization, signal processing, and reinforcement learning communities.

References in corpus (3)

• Constrained Consensus
• Harnessing Smoothness to Accelerate Distributed Optimization
• $QD$-Learning: A Collaborative Distributed Strategy for Multi-Agent Reinforcement Learning Through Consensus + Innovations

Cited by in corpus (5)

• Pervasive AI for IoT applications: A Survey on Resource-efficient Distributed Artificial Intelligence
• Deep Reinforcement Learning for Radio Resource Allocation and Management in Next Generation Heterogeneous Wireless Networks: A Survey
• Byzantine-Resilient Decentralized TD Learning with Linear Function Approximation
• A Survey on Agent Workflow -- Status and Future
• Task-Oriented Data Compression for Multi-Agent Communications Over Bit-Budgeted Channels