arrow
Return

Scalable and Sample Efficient Distributed Policy Gradient Algorithms in Multi-Agent Networked Systems

delete2026-01-01
delete0
PRE
AI
X
Xin Liu
H
Honghao Wei
L
Lei Ying
DOI:10.1109/TON.2025.3647225delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
This paper studies a class of multi-agent reinforcement learning (MARL) problems where the reward that an agent receives depends on the states of other agents, but the next state only depends on the agent’s own current state and action. We name it REC-MARL standing for REward-Coupled Multi-Agent Reinforcement Learning. REC-MARL has a range of important applications such as real-time access control and distributed power control in wireless networks. This paper presents a distributed policy gradient algorithm for REC-MARL. The proposed algorithm is distributed in two aspects: 1) the learned policy is a distributed policy that maps a local state of an agent to its local action and 2) the learning/training is distributed, during which each agent updates its policy based on its own and neighbors’ information. The learned algorithm achieves a stationary policy and its iterative complexity bounds depend on the dimension of local states and actions. The experimental results of our algorithm for the real-time access control and power control in wireless networks show that our policy significantly outperforms the state-of-the-art algorithms and well-known benchmarks.
Keywords:
Multi-agent networked systems
distributed reinforcement learning
real-time access control
wireless power control

Journal

I
IEEE Transactions on Networking
IF:
0
Papers:
543
Citations:
0

Organization

S
Shanghaitech University
Scholars:
666
Papers: 201
Citations: 0
W
washington state university
Scholars:
1.8W
Papers: 1.6W
Citations: 114
U
university of michigan
Scholars:
8.8K
Papers: 4.2K
Citations: 1
researcher View more organizations