Saved in:
Bibliographic Details
Main Authors: Bochem, Severin, Gonzalez-Sanchez, Eduardo, Bicker, Yves, Fadini, Gabriele
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2411.19732
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929608897593344
author Bochem, Severin
Gonzalez-Sanchez, Eduardo
Bicker, Yves
Fadini, Gabriele
author_facet Bochem, Severin
Gonzalez-Sanchez, Eduardo
Bicker, Yves
Fadini, Gabriele
contents Reinforcement learning often requires extensive training data. Simulation-to-real transfer offers a promising approach to address this challenge in robotics. While differentiable simulators offer improved sample efficiency through exact gradients, they can be unstable in contact-rich environments and may lead to poor generalization. This paper introduces a novel approach integrating sharpness-aware optimization into gradient-based reinforcement learning algorithms. Our simulation results demonstrate that our method, tested on contact-rich environments, significantly enhances policy robustness to environmental variations and action perturbations while maintaining the sample efficiency of first-order methods. Specifically, our approach improves action noise tolerance compared to standard first-order methods and achieves generalization comparable to zeroth-order methods. This improvement stems from finding flatter minima in the loss landscape, associated with better generalization. Our work offers a promising solution to balance efficient learning and robust sim-to-real transfer in robotics, potentially bridging the gap between simulation and real-world performance.
format Preprint
id arxiv_https___arxiv_org_abs_2411_19732
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Improving generalization of robot locomotion policies via Sharpness-Aware Reinforcement Learning
Bochem, Severin
Gonzalez-Sanchez, Eduardo
Bicker, Yves
Fadini, Gabriele
Robotics
Artificial Intelligence
Machine Learning
Reinforcement learning often requires extensive training data. Simulation-to-real transfer offers a promising approach to address this challenge in robotics. While differentiable simulators offer improved sample efficiency through exact gradients, they can be unstable in contact-rich environments and may lead to poor generalization. This paper introduces a novel approach integrating sharpness-aware optimization into gradient-based reinforcement learning algorithms. Our simulation results demonstrate that our method, tested on contact-rich environments, significantly enhances policy robustness to environmental variations and action perturbations while maintaining the sample efficiency of first-order methods. Specifically, our approach improves action noise tolerance compared to standard first-order methods and achieves generalization comparable to zeroth-order methods. This improvement stems from finding flatter minima in the loss landscape, associated with better generalization. Our work offers a promising solution to balance efficient learning and robust sim-to-real transfer in robotics, potentially bridging the gap between simulation and real-world performance.
title Improving generalization of robot locomotion policies via Sharpness-Aware Reinforcement Learning
topic Robotics
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2411.19732