TY - JOUR AU - Cohen, Andrew AU - Qiao, Xingye AU - Yu, Lei AU - Way, Elliot AU - Tong, Xiangrong PY - 2019/07/17 Y2 - 2024/03/29 TI - Diverse Exploration via Conjugate Policies for Policy Gradient Methods JF - Proceedings of the AAAI Conference on Artificial Intelligence JA - AAAI VL - 33 IS - 01 SE - AAAI Technical Track: Machine Learning DO - 10.1609/aaai.v33i01.33013404 UR - https://ojs.aaai.org/index.php/AAAI/article/view/4215 SP - 3404-3411 AB - <p>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.</p> ER -