DODT: Enhanced Online Decision Transformer Learning through Dreamer's Actor-Critic Trajectory Forecasting

TL;DR

This paper presents DODT, a novel reinforcement learning framework that combines Dreamer's trajectory forecasting with decision transformers, leading to improved sample efficiency and robustness in complex decision-making tasks.

Contribution

The paper introduces an integrated approach that merges Dreamer's world model with decision transformers, enabling bidirectional learning and enhanced performance in reinforcement learning.

Findings

Achieves better sample efficiency than existing methods.

Demonstrates improved reward maximization across benchmarks.

Shows robustness in diverse, dynamic environments.

Abstract

Advancements in reinforcement learning have led to the development of sophisticated models capable of learning complex decision-making tasks. However, efficiently integrating world models with decision transformers remains a challenge. In this paper, we introduce a novel approach that combines the Dreamer algorithm's ability to generate anticipatory trajectories with the adaptive learning strengths of the Online Decision Transformer. Our methodology enables parallel training where Dreamer-produced trajectories enhance the contextual decision-making of the transformer, creating a bidirectional enhancement loop. We empirically demonstrate the efficacy of our approach on a suite of challenging benchmarks, achieving notable improvements in sample efficiency and reward maximization over existing methods. Our results indicate that the proposed integrated framework not only accelerates learning but also showcases robustness in diverse and dynamic scenarios, marking a significant step forward in model-based reinforcement learning.