Mitigating Suboptimality of Deterministic Policy Gradients in Complex Q-functions

Jain, Ayush, Kosaka, Norio, Li, Xinhu, Kim, Kyung-Min, Bıyık, Erdem, Lim, Joseph J.

arXiv.org Machine Learning 

In reinforcement learning, off-policy actor-critic approaches like DDPG and TD3 are based on the deterministic policy gradient. Herein, the Q-function is trained from off-policy environment data and the actor (policy) is trained to maximize the Q-function via gradient ascent. We observe that in complex tasks like dexterous manipulation and restricted locomotion, the Q-value is a complex function of action, having several local optima or discontinuities. This poses a challenge for gradient ascent to traverse and makes the actor prone to get stuck at local optima. To address this, we introduce a new actor architecture that combines two simple insights: (i) use multiple actors and evaluate the Q-value maximizing action, and (ii) learn surrogates to the Q-function that are simpler to optimize with gradient-based methods. We evaluate tasks such as restricted locomotion, dexterous manipulation, and large discrete-action space recommender systems and show that our actor finds optimal actions more frequently and outperforms alternate actor architectures.

Duplicate Docs Excel Report

Title
None found

Similar Docs  Excel Report  more

TitleSimilaritySource
None found