A Scalable Finite Difference Method for Deep Reinforcement Learning

Matthew W Allen, John C. Raisbeck, Hakho Lee

Keywords: Deep Reinforcement Learning, Reinforcement Learning, Finite Differences, Distributed Systems, Policy Optimization

TL;DR: An efficient & scalable method for policy optimization using finite differences and a modification of SGD.

Abstract: Several low-bandwidth distributable black-box optimization algorithms have recently been shown to perform nearly as well as more refined modern methods in some Deep Reinforcement Learning domains. In this work we investigate a core problem with the use of distributed workers in such systems. Further, we investigate the dramatic differences in performance between the popular Adam gradient descent algorithm and the simplest form of stochastic gradient descent. These investigations produce a stable, low-bandwidth learning algorithm that achieves 100\% usage of all connected CPUs under typical conditions.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Reinforcement Learning (eg, decision and control, planning, hierarchical RL, robotics)