Optimizing Robotic Manipulation with Decision-RWKV: A Recurrent Sequence Modeling Approach for Lifelong Learning

TL;DR

This paper introduces Decision-RWKV, a novel sequence modeling approach that enhances lifelong robotic learning by integrating the RWKV framework with decision transformers, demonstrating improved performance in sequential decision tasks.

Contribution

The paper presents the Decision-RWKV model, combining RWKV with decision transformers and experience replay to advance lifelong learning in robotics, a novel integration not previously explored.

Findings

Decision-RWKV outperforms traditional models in lifelong learning tasks.

The model effectively handles multiple subtasks in robotic environments.

Open-source code facilitates reproducibility and further research.

Abstract

Models based on the Transformer architecture have seen widespread application across fields such as natural language processing, computer vision, and robotics, with large language models like ChatGPT revolutionizing machine understanding of human language and demonstrating impressive memory and reproduction capabilities. Traditional machine learning algorithms struggle with catastrophic forgetting, which is detrimental to the diverse and generalized abilities required for robotic deployment. This paper investigates the Receptance Weighted Key Value (RWKV) framework, known for its advanced capabilities in efficient and effective sequence modeling, and its integration with the decision transformer and experience replay architectures. It focuses on potential performance enhancements in sequence decision-making and lifelong robotic learning tasks. We introduce the Decision-RWKV (DRWKV) model and conduct extensive experiments using the D4RL database within the OpenAI Gym environment and on the D'Claw platform to assess the DRWKV model's performance in single-task tests and lifelong learning scenarios, showcasing its ability to handle multiple subtasks efficiently. The code for all algorithms, training, and image rendering in this study is open-sourced at https://github.com/ancorasir/DecisionRWKV.