Task-Aware Exploration via a Predictive Bisimulation Metric
This work addresses the problem of efficient exploration in visual domains for reinforcement learning researchers, offering an incremental improvement by mitigating representation collapse and enhancing task-aware strategies.
The paper tackles the challenge of accelerating exploration in visual reinforcement learning under sparse rewards by introducing TEB, a task-aware exploration approach that uses a predictive bisimulation metric to learn task-relevant representations and measure novelty, achieving superior exploration ability and outperforming recent baselines in experiments on MetaWorld and Maze2D.
Accelerating exploration in visual reinforcement learning under sparse rewards remains challenging due to the substantial task-irrelevant variations. Despite advances in intrinsic exploration, many methods either assume access to low-dimensional states or lack task-aware exploration strategies, thereby rendering them fragile in visual domains. To bridge this gap, we present TEB, a Task-aware Exploration approach that tightly couples task-relevant representations with exploration through a predictive Bisimulation metric. Specifically, TEB leverages the metric not only to learn behaviorally grounded task representations but also to measure behaviorally intrinsic novelty over the learned latent space. To realize this, we first theoretically mitigate the representation collapse of degenerate bisimulation metrics under sparse rewards by internally introducing a simple but effective predicted reward differential. Building on this robust metric, we design potential-based exploration bonuses, which measure the relative novelty of adjacent observations over the latent space. Extensive experiments on MetaWorld and Maze2D show that TEB achieves superior exploration ability and outperforms recent baselines.