Diverse Exploration via Conjugate Policies for Policy Gradient Methods
DOI:
https://doi.org/10.1609/aaai.v33i01.33013404Abstract
We address the challenge of effective exploration while maintaining good performance in policy gradient methods. As a solution, we propose diverse exploration (DE) via conjugate policies. DE learns and deploys a set of conjugate policies which can be conveniently generated as a byproduct of conjugate gradient descent. We provide both theoretical and empirical results showing the effectiveness of DE at achieving exploration, improving policy performance, and the advantage of DE over exploration by random policy perturbations.
Downloads
Published
2019-07-17
How to Cite
Cohen, A., Qiao, X., Yu, L., Way, E., & Tong, X. (2019). Diverse Exploration via Conjugate Policies for Policy Gradient Methods. Proceedings of the AAAI Conference on Artificial Intelligence, 33(01), 3404-3411. https://doi.org/10.1609/aaai.v33i01.33013404
Issue
Section
AAAI Technical Track: Machine Learning