Exclusively Penalized Q-learning for Offline Reinforcement Learning
This addresses a specific limitation in offline RL methods for researchers and practitioners, but it is incremental as it builds on existing penalized value function approaches.
The paper tackles the underestimation bias in offline reinforcement learning by proposing Exclusively Penalized Q-learning (EPQ), which selectively penalizes states prone to errors, resulting in significant reduction of bias and improved performance in offline control tasks.
Constraint-based offline reinforcement learning (RL) involves policy constraints or imposing penalties on the value function to mitigate overestimation errors caused by distributional shift. This paper focuses on a limitation in existing offline RL methods with penalized value function, indicating the potential for underestimation bias due to unnecessary bias introduced in the value function. To address this concern, we propose Exclusively Penalized Q-learning (EPQ), which reduces estimation bias in the value function by selectively penalizing states that are prone to inducing estimation errors. Numerical results show that our method significantly reduces underestimation bias and improves performance in various offline control tasks compared to other offline RL methods