Can Reinforcement Learning for Continuous Control Generalize Across Physics Engines?

Abstract

Reinforcement learning (RL) algorithms should learn as much as possible about the environment but not the properties of the physics engines that generate the environment. There are multiple algorithms that solve the task in a physics engine based environment but there is no work done so far to understand if the RL algorithms can generalize across physics engines. In this work, we compare the generalization performance of various deep reinforcement learning algorithms on a variety of control tasks. Our results show that MuJoCo is the best engine to transfer the learning to other engines. On the other hand, none of the algorithms generalize when trained on PyBullet. We also found out that various algorithms have a promising generalizability if the effect of random seeds can be minimized on their performance.

arXiv comment10 pages plus appendix

References in corpus (6)

• Learning agile and dynamic motor skills for legged robots
• Benchmarking Deep Reinforcement Learning for Continuous Control
• Challenges of Real-World Reinforcement Learning
• Measuring the Reliability of Reinforcement Learning Algorithms
• Benchmarking Simulated Robotic Manipulation through a Real World Dataset
• Traversing the Reality Gap via Simulator Tuning