TY - JOUR AU - Tang, Yunhao AU - Agrawal, Shipra PY - 2020/04/03 Y2 - 2024/03/28 TI - Discretizing Continuous Action Space for On-Policy Optimization JF - Proceedings of the AAAI Conference on Artificial Intelligence JA - AAAI VL - 34 IS - 04 SE - AAAI Technical Track: Machine Learning DO - 10.1609/aaai.v34i04.6059 UR - https://ojs.aaai.org/index.php/AAAI/article/view/6059 SP - 5981-5988 AB - <p>In this work, we show that discretizing action space for continuous control is a simple yet powerful technique for on-policy optimization. The explosion in the number of discrete actions can be efficiently addressed by a policy with factorized distribution across action dimensions. We show that the discrete policy achieves significant performance gains with state-of-the-art on-policy optimization algorithms (PPO, TRPO, ACKTR) especially on high-dimensional tasks with complex dynamics. Additionally, we show that an ordinal parameterization of the discrete distribution can introduce the inductive bias that encodes the natural ordering between discrete actions. This ordinal architecture further significantly improves the performance of PPO/TRPO.</p> ER -