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.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keymohammed2020can

Related papers

• Exploring Reinforcement Learning Techniques For Discrete And Continuous Control Tasks In The Mujoco Environment (2023)0.00
• Generalization In Transfer Learning (2019)0.00
• Learning Robust And Adaptive Real-world Continuous Control Using Simulation And Transfer Learning (2018)0.00
• Knowledge Transfer In Multi-task Deep Reinforcement Learning For Continuous Control (2020)0.00
• Assessing Generalization In Deep Reinforcement Learning (2018)0.00
• Trajectory-wise Multiple Choice Learning For Dynamics Generalization In Reinforcement Learning (2020)6.77
• Towards An Adaptable And Generalizable Optimization Engine In Decision And Control: A Meta Reinforcement Learning Approach (2024)0.00
• Illuminating Generalization In Deep Reinforcement Learning Through Procedural Level Generation (2018)0.00