Abstract
Effective exploration in reinforcement learning requires not only tracking where an agent has been, but also understanding how the agent perceives and represents the world. To learn powerful representations, an agent should actively explore states that contribute to its knowledge of the environment. Temporal representations can capture the information necessary to solve a wide range of potential tasks while avoiding the computational cost associated with full state reconstruction. In this paper, we propose an exploration method that leverages temporal contrastive representations to guide exploration, prioritizing states with unpredictable future outcomes. We demonstrate that such representations can enable the learning of complex exploratory x in locomotion, manipulation, and embodied-AI tasks, revealing capabilities and behaviors that traditionally require extrinsic rewards. Unlike approaches that rely on explicit distance learning or episodic memory mechanisms (e.g., quasimetric-based methods), our method builds directly on temporal similarities, yielding a simpler yet effective strategy for exploration.
Overview
The agent's starting state is $( s_0 )$. We train a contrastive model such that the temporal similarity between the representation of $((s_0, a_0))$ and $( s_{2,3,4,\ldots})$ is high, and we reward the agent for visiting states that are further in the future. For example, the reward for visiting $( s_4 )$ from $( s_0 )$ should be larger than the reward for visiting $( s_3 )$ from the same $( s_0 )$.
Reward Visualization
Evolution of the C-TeC reward during training. This figure shows how the intrinsic reward changes over the course of training based on future state visitation. The black circle in the lower-left corner represents the starting state. Early in training (3M steps), higher rewards are assigned to nearby states. As training progresses, the agent explores farther, and the reward increases for more distant regions. All reward values are normalized for visualization.
