Optimal Policy Sparsification and Low Rank Decomposition for Deep Reinforcement Learning

arXiv:2403.06313v1 Announce Type: new
Abstract: Deep reinforcement learning(DRL) has shown significant promise in a wide range of applications including computer games and robotics. Yet, training DRL policies consume extraordinary computing resources resulting in dense policies which are prone to overfitting. Moreover, inference with dense DRL policies limit their practical applications, especially in edge computing. Techniques such as pruning and singular value decomposition have been used with deep learning models to achieve sparsification and model compression to limit overfitting and reduce …

abstract applications arxiv computer computing computing resources cs.ai cs.lg games inference low overfitting policy practical reinforcement reinforcement learning resources robotics training type