Keywords: model-based reinforcement learning, visual reinforcement learning, pretraining
TL;DR: We investigate the feasibility of pretraining and cross-task transfer in model-based RL, and improve sample-efficiency substantially over baselines on the Atari100k benchmark.
Abstract: Reinforcement Learning (RL) algorithms can solve challenging control problems directly from image observations, but they often require millions of environment interactions to do so. Recently, model-based RL algorithms have greatly improved sample-efficiency by concurrently learning an internal model of the world, and supplementing real environment interactions with imagined rollouts for policy improvement. However, learning an effective model of the world from scratch is challenging, and in stark contrast to humans that rely heavily on world understanding and visual cues for learning new skills. In this work, we investigate whether internal models learned by modern model-based RL algorithms can be leveraged to solve new, distinctly different tasks faster. We propose Model-Based Cross-Task Transfer (XTRA), a framework for sample-efficient online RL with scalable pretraining and finetuning of learned world models. By offline multi-task pretraining and online cross-task finetuning, we achieve substantial improvements on the Atari100k benchmark over a baseline trained from scratch; we improve mean performance of model-based algorithm EfficientZero by 23%, and by as much as 71% in some instances. Project page: https://nicklashansen.github.io/xtra
Community Implementations: [![CatalyzeX](/images/catalyzex_icon.svg) 1 code implementation](https://www.catalyzex.com/paper/arxiv:2210.10763/code)