Addressing stale gradients in asynchronous federated deep reinforcement learning

Advancements in reinforcement learning (RL) via deep neural networks have enabled their application to a variety of real-world problems. However, these applications often suffer from long training times. While attempts to distribute training have been successful in controlled scenarios, they face challenges in heterogeneous-capacity, unstable, and privacy critical environments. This work applies concepts from federated learning (FL) to distributed RL, specifically addressing the stale gradient problem. A deterministic framework for asynchronous federated RL is utilized to explore dynamic methods for handling stale gradient updates in the Arcade Learning Environment. Experimental results from applying these methods to two Atari-2600 games demonstrate a relative speedup of up to 95\% compared to plain A3C in large and unstable federations.

Name thesis:degree_name

Master of Science

Level thesis:degree_level

thesis

Discipline thesis:degree_discipline

Computer science

Department dc:contributor.department

Department of Computer Science

Grantor

Iowa State University

Year dc:date.issued

2023

Author dc:creator

• Stanley, Justin

Advisors dc:contributor.advisor

• Jannesari, Ali
• Quinn, Christopher
• Tian, Jin
• Huai, Mengdi

Language dc:language.iso

en

DOI

https://doi.org/10.31274/td-20240329-43

OAI identifier oai:identifier

oai:dr.lib.iastate.edu:20.500.12876/YvkAO7Bz