Fully asynchronous policy evaluation in distributed reinforcement learning over networks

This paper proposes a fully asynchronous scheme for the policy evaluation problem of distributed reinforcement learning (DisRL) over directed peer-to-peer networks. Without waiting for any other node of the network, each node can locally update its value function at any time using (possibly delayed)...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Automatica (Oxford) Jg. 136; S. 110092
Hauptverfasser: Sha, Xingyu, Zhang, Jiaqi, You, Keyou, Zhang, Kaiqing, Başar, Tamer
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier Ltd 01.02.2022
Schlagworte:
Distributed reinforcement learning
Fully asynchronous updates
Multi-agent networks
Policy evaluation
Policy evaluation
Multi-agent networks
Distributed reinforcement learning
Fully asynchronous updates
ISSN:0005-1098, 1873-2836
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:This paper proposes a fully asynchronous scheme for the policy evaluation problem of distributed reinforcement learning (DisRL) over directed peer-to-peer networks. Without waiting for any other node of the network, each node can locally update its value function at any time using (possibly delayed) information from its neighbors. This is in sharp contrast to the gossip-based scheme where a pair of nodes concurrently update. Even though the fully asynchronous setting involves a difficult multi-timescale decision problem, we design a novel incremental aggregated gradient (IAG) based distributed algorithm and develop a push–pull augmented graph approach to prove its exact convergence at a linear rate of O(ck) where c∈(0,1) and k is the total number of updates within the entire network. Finally, numerical experiments validate that our method speeds up linearly with respect to the number of nodes, and is robust to straggler nodes.
ISSN:0005-1098
1873-2836
DOI:10.1016/j.automatica.2021.110092