Distributed Reinforcement Learning via Gossip

DOI10.1109/TAC.2016.2585302MaRDI QIDQ5282396zbMATH OpenOpenAlexFDO

Authors Adwaitvedant S. Mathkar, Vivek Borkar

Publication date 27 July 2017

Published in IEEE Transactions on Automatic Control (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1310.7610

Mathematics Subject Classification ID

Markov processes: estimation; hidden Markov models (62M05) Learning and adaptive systems in artificial intelligence (68T05) Social networks; opinion dynamics (91D30) Distributed algorithms (68W15)

Abstract: We consider the classical TD(0) algorithm implemented on a network of agents wherein the agents also incorporate the updates received from neighboring agents using a gossip-like mechanism. The combined scheme is shown to converge for both discounted and average cost problems.

Cited in

(7)

• Distributed consensus-based multi-agent temporal-difference learning
• Decentralized fused-learner architectures for Bayesian reinforcement learning
• Distributed Deep Learning on Heterogeneous Computing Resources Using Gossip Communication
• Scalable Reinforcement Learning for Multiagent Networked Systems
• Distributed L₂-gain control of large-scale systems under gossip communication protocol
• Model-free adaptive control design for nonlinear discrete-time processes with reinforcement learning techniques
• Finite-time performance of distributed temporal-difference learning with linear function approximation

This page was built for publication: Distributed Reinforcement Learning via Gossip

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5282396)