Introspective Experience Replay: Look Back When Surprised
This addresses a specific problem in reinforcement learning for improving training stability and efficiency, representing an incremental advancement over prior techniques.
The paper tackles the sub-optimal convergence and high seed sensitivity in reinforcement learning experience replay methods by proposing Introspective Experience Replay (IER), which selectively samples data before surprising events, and demonstrates reliable and superior performance compared to existing methods across most tasks.
In reinforcement learning (RL), experience replay-based sampling techniques play a crucial role in promoting convergence by eliminating spurious correlations. However, widely used methods such as uniform experience replay (UER) and prioritized experience replay (PER) have been shown to have sub-optimal convergence and high seed sensitivity respectively. To address these issues, we propose a novel approach called IntrospectiveExperience Replay (IER) that selectively samples batches of data points prior to surprising events. Our method builds upon the theoretically sound reverse experience replay (RER) technique, which has been shown to reduce bias in the output of Q-learning-type algorithms with linear function approximation. However, this approach is not always practical or reliable when using neural function approximation. Through empirical evaluations, we demonstrate that IER with neural function approximation yields reliable and superior performance compared toUER, PER, and hindsight experience replay (HER) across most tasks.