Temporal Distance-aware Transition Augmentation for Offline Model-based Reinforcement Learning

TL;DR

This paper introduces TempDATA, a novel offline model-based reinforcement learning framework that generates temporally structured transition augmentations in latent space, improving performance in long-horizon, sparse-reward tasks.

Contribution

TempDATA is the first to model temporal distance in latent space for transition augmentation, enhancing offline MBRL in complex long-horizon environments.

Findings

TempDATA outperforms previous offline MBRL methods.

Achieves comparable or better results than diffusion-based augmentation.

Effective in long-horizon, sparse-reward tasks.

Abstract

The goal of offline reinforcement learning (RL) is to extract a high-performance policy from the fixed datasets, minimizing performance degradation due to out-of-distribution (OOD) samples. Offline model-based RL (MBRL) is a promising approach that ameliorates OOD issues by enriching state-action transitions with augmentations synthesized via a learned dynamics model. Unfortunately, seminal offline MBRL methods often struggle in sparse-reward, long-horizon tasks. In this work, we introduce a novel MBRL framework, dubbed Temporal Distance-Aware Transition Augmentation (TempDATA), that generates augmented transitions in a temporally structured latent space rather than in raw state space. To model long-horizon behavior, TempDATA learns a latent abstraction that captures a temporal distance from both trajectory and transition levels of state space. Our experiments confirm that TempDATA outperforms previous offline MBRL methods and achieves matching or surpassing the performance of diffusion-based trajectory augmentation and goal-conditioned RL on the D4RL AntMaze, FrankaKitchen, CALVIN, and pixel-based FrankaKitchen.