Episodic Novelty Through Temporal Distance
This addresses the problem of inefficient exploration in CMDPs for reinforcement learning practitioners, offering a novel method that improves upon existing incremental approaches.
The paper tackles the challenge of exploration in sparse reward environments, particularly in Contextual Markov Decision Processes (CMDPs), by proposing Episodic Novelty Through Temporal Distance (ETD), which uses temporal distance as a metric for intrinsic rewards and significantly outperforms state-of-the-art methods in benchmark tasks.
Exploration in sparse reward environments remains a significant challenge in reinforcement learning, particularly in Contextual Markov Decision Processes (CMDPs), where environments differ across episodes. Existing episodic intrinsic motivation methods for CMDPs primarily rely on count-based approaches, which are ineffective in large state spaces, or on similarity-based methods that lack appropriate metrics for state comparison. To address these shortcomings, we propose Episodic Novelty Through Temporal Distance (ETD), a novel approach that introduces temporal distance as a robust metric for state similarity and intrinsic reward computation. By employing contrastive learning, ETD accurately estimates temporal distances and derives intrinsic rewards based on the novelty of states within the current episode. Extensive experiments on various benchmark tasks demonstrate that ETD significantly outperforms state-of-the-art methods, highlighting its effectiveness in enhancing exploration in sparse reward CMDPs.