Saved in:
Bibliographic Details
Main Authors: Shin, Jaeyong, Cha, Woohyun, Kim, Donghyeon, Cha, Junhyeok, Park, Jaeheung
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.08246
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915876942381056
author Shin, Jaeyong
Cha, Woohyun
Kim, Donghyeon
Cha, Junhyeok
Park, Jaeheung
author_facet Shin, Jaeyong
Cha, Woohyun
Kim, Donghyeon
Cha, Junhyeok
Park, Jaeheung
contents Reinforcement learning (RL) has shown great potential in training agile and adaptable controllers for legged robots, enabling them to learn complex locomotion behaviors directly from experience. However, policies trained in simulation often fail to transfer to real-world robots due to unrealistic assumptions such as infinite actuator bandwidth and the absence of torque limits. These conditions allow policies to rely on abrupt, high-frequency torque changes, which are infeasible for real actuators with finite bandwidth. Traditional methods address this issue by penalizing aggressive motions through regularization rewards, such as joint velocities, accelerations, and energy consumption, but they require extensive hyperparameter tuning. Alternatively, Lipschitz-Constrained Policies (LCP) enforce finite bandwidth action control by penalizing policy gradients, but their reliance on gradient calculations introduces significant GPU memory overhead. To overcome this limitation, this work proposes Spectral Normalization (SN) as an efficient replacement for enforcing Lipschitz continuity. By constraining the spectral norm of network weights, SN effectively limits high-frequency policy fluctuations while significantly reducing GPU memory usage. Experimental evaluations in both simulation and real-world humanoid robot show that SN achieves performance comparable to gradient penalty methods while enabling more efficient parallel training.
format Preprint
id arxiv_https___arxiv_org_abs_2504_08246
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Spectral Normalization for Lipschitz-Constrained Policies on Learning Humanoid Locomotion
Shin, Jaeyong
Cha, Woohyun
Kim, Donghyeon
Cha, Junhyeok
Park, Jaeheung
Robotics
Machine Learning
Systems and Control
Reinforcement learning (RL) has shown great potential in training agile and adaptable controllers for legged robots, enabling them to learn complex locomotion behaviors directly from experience. However, policies trained in simulation often fail to transfer to real-world robots due to unrealistic assumptions such as infinite actuator bandwidth and the absence of torque limits. These conditions allow policies to rely on abrupt, high-frequency torque changes, which are infeasible for real actuators with finite bandwidth. Traditional methods address this issue by penalizing aggressive motions through regularization rewards, such as joint velocities, accelerations, and energy consumption, but they require extensive hyperparameter tuning. Alternatively, Lipschitz-Constrained Policies (LCP) enforce finite bandwidth action control by penalizing policy gradients, but their reliance on gradient calculations introduces significant GPU memory overhead. To overcome this limitation, this work proposes Spectral Normalization (SN) as an efficient replacement for enforcing Lipschitz continuity. By constraining the spectral norm of network weights, SN effectively limits high-frequency policy fluctuations while significantly reducing GPU memory usage. Experimental evaluations in both simulation and real-world humanoid robot show that SN achieves performance comparable to gradient penalty methods while enabling more efficient parallel training.
title Spectral Normalization for Lipschitz-Constrained Policies on Learning Humanoid Locomotion
topic Robotics
Machine Learning
Systems and Control
url https://arxiv.org/abs/2504.08246