Pruning With Scaled Policy Constraints for Light-Weight Reinforcement Learning

Pruning With Scaled Policy Constraints for Light-Weight Reinforcement Learning

The increasing computational demands of Deep Reinforcement Learning (DRL) models, particularly for embedded systems in autonomous vehicles and drones, present significant challenges owing to their extensive neural network complexities. Previous DRL compression strategies predominantly focused on uns...

Full description

Bibliographic Details
Main Authors: Seongmin Park, Hyungmin Kim, Hyunhak Kim, Jungwook Choi
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Deep reinforcement learning
offline reinforcement learning
model compression
structured pruning
fine-tuning
robotics
Online Access:https://ieeexplore.ieee.org/document/10439169/

Internet

https://ieeexplore.ieee.org/document/10439169/

Similar Items