MOTO: Offline Pre-training to Online Fine-tuning for Model-based Robot Learning

Rafael Rafailov, Kyle Beltran Hatch, Victor Kolev, John D Martin, Mariano Phielipp, Chelsea Finn

Keywords: offline RL, online fine-tuning, model-learning, robot learning

TL;DR: We develop a model-based RL method specifically designed for online fine-tuning of robot tasks. MOTO is the first method to solve the Franka Kitchen environment from images.

Abstract: We study the problem of offline pre-training and online fine-tuning for reinforcement learning from high-dimensional observations in the context of realistic robot tasks. Recent offline model-free approaches successfully use online fine-tuning to either improve the performance of the agent over the data collection policy or adapt to novel tasks. At the same time, model-based RL algorithms have achieved significant progress in sample efficiency and the complexity of the tasks they can solve, yet remain under-utilized in the fine-tuning setting. In this work, we argue that existing methods for high-dimensional model-based offline RL are not suitable for offline-to-online fine-tuning due to issues with distribution shifts, off-dynamics data, and non-stationary rewards. We propose an on-policy model-based method that can efficiently reuse prior data through model-based value expansion and policy regularization, while preventing model exploitation by controlling epistemic uncertainty. We find that our approach successfully solves tasks from the MetaWorld benchmark, as well as the Franka Kitchen robot manipulation environment completely from images. To our knowledge, MOTO is the first and only method to solve this environment from pixels.

Student First Author: yes

Instructions: I have read the instructions for authors (https://corl2023.org/instructions-for-authors/)

Website: https://sites.google.com/view/mo2o/