Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learning

by   Tianhe Yu, et al.

Meta-reinforcement learning algorithms can enable robots to acquire new skills much more quickly, by leveraging prior experience to learn how to learn. However, much of the current research on meta-reinforcement learning focuses on task distributions that are very narrow. For example, a commonly used meta-reinforcement learning benchmark uses different running velocities for a simulated robot as different tasks. When policies are meta-trained on such narrow task distributions, they cannot possibly generalize to more quickly acquire entirely new tasks. Therefore, if the aim of these methods is to enable faster acquisition of entirely new behaviors, we must evaluate them on task distributions that are sufficiently broad to enable generalization to new behaviors. In this paper, we propose an open-source simulated benchmark for meta-reinforcement learning and multi-task learning consisting of 50 distinct robotic manipulation tasks. Our aim is to make it possible to develop algorithms that generalize to accelerate the acquisition of entirely new, held-out tasks. We evaluate 6 state-of-the-art meta-reinforcement learning and multi-task learning algorithms on these tasks. Surprisingly, while each task and its variations (e.g., with different object positions) can be learned with reasonable success, these algorithms struggle to learn with multiple tasks at the same time, even with as few as ten distinct training tasks. Our analysis and open-source environments pave the way for future research in multi-task learning and meta-learning that can enable meaningful generalization, thereby unlocking the full potential of these methods.


page 2

page 5

page 13


Unsupervised Meta-Learning for Reinforcement Learning

Meta-learning is a powerful tool that builds on multi-task learning to l...

MELD: Meta-Reinforcement Learning from Images via Latent State Models

Meta-reinforcement learning algorithms can enable autonomous agents, suc...

Generalization in Dexterous Manipulation via Geometry-Aware Multi-Task Learning

Dexterous manipulation of arbitrary objects, a fundamental daily task fo...

MT-Opt: Continuous Multi-Task Robotic Reinforcement Learning at Scale

General-purpose robotic systems must master a large repertoire of divers...

On the Effectiveness of Fine-tuning Versus Meta-reinforcement Learning

Intelligent agents should have the ability to leverage knowledge from pr...

CompoSuite: A Compositional Reinforcement Learning Benchmark

We present CompoSuite, an open-source simulated robotic manipulation ben...

Demonstration-Bootstrapped Autonomous Practicing via Multi-Task Reinforcement Learning

Reinforcement learning systems have the potential to enable continuous i...

Please sign up or login with your details

Forgot password? Click here to reset