GTA: Generative Trajectory Augmentation with Guidance for Offline Reinforcement Learning

TL;DR

GTA introduces a diffusion-based generative data augmentation method that enhances offline RL by creating high-reward, plausible trajectories, leading to improved policy performance and data quality across diverse tasks.

Contribution

The paper presents GTA, a novel diffusion model-based trajectory augmentation technique that directly improves offline RL data quality and policy outcomes.

Findings

GTA improves offline RL performance across various tasks.

GTA enhances the quality of augmented data.

GTA outperforms existing augmentation methods.

Abstract

Offline Reinforcement Learning (Offline RL) presents challenges of learning effective decision-making policies from static datasets without any online interactions. Data augmentation techniques, such as noise injection and data synthesizing, aim to improve Q-function approximation by smoothing the learned state-action region. However, these methods often fall short of directly improving the quality of offline datasets, leading to suboptimal results. In response, we introduce GTA, Generative Trajectory Augmentation, a novel generative data augmentation approach designed to enrich offline data by augmenting trajectories to be both high-rewarding and dynamically plausible. GTA applies a diffusion model within the data augmentation framework. GTA partially noises original trajectories and then denoises them with classifier-free guidance via conditioning on amplified return value. Our results show that GTA, as a general data augmentation strategy, enhances the performance of widely used offline RL algorithms across various tasks with unique challenges. Furthermore, we conduct a quality analysis of data augmented by GTA and demonstrate that GTA improves the quality of the data. Our code is available at https://github.com/Jaewoopudding/GTA