Work

In value-based deep reinforcement learning, a pruned network is a good network

Year: 2024

Type: preprint

Abstract: Recent work has shown that deep reinforcement learning agents have difficulty in effectively using their network parameters. We leverage prior insights into the advantages of sparse training techniques and demonstrate that gradual magnitude pruning enables value-based agents to maximize parameter effectiveness. This results in networks that yield dramatic performance improvements over traditional networks, using only a small fraction of the full network parameters.

Source: arXiv (Cornell University)

DOI: 10.48550/arxiv.2402.12479

Authors: Obando-Ceron, Johan, Courville, Aaron, Castro, Pablo Samuel

Institutions: Mila - Quebec Artificial Intelligence Institute, Université de Montréal

Language: English

Topic: Reinforcement Learning in Robotics

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funder: DeepMind

OpenAlex ID: W4392019712

Cites:

Cited by:

Related to: 10