vix.ing · top · new · best · stats

Honey, I Shrunk The Actor: A Case Study on Preserving Performance with Smaller Actors in Actor-Critic RL

2021/02/23 by Siddharth Mysore, Mysore, Siddharth, Bassel Mabsout +5 · 4 citations
Agricultural and Biological Sciences · Computer Science · Engineering · #Advancements in Semiconductor Devices and Circuit Design #FOS: Computer and information sciences #Insect symbiosis and bacterial influences #Machine Learning (cs.LG) #Neural Networks and Reservoir Computing #Robotics (cs.RO) #cs.LG #cs.RO

paper · pdf · doi:10.48550/arxiv.2102.11893

Accepted to the IEEE Conference on Games 2021

openalex publication_date 2021/02/23 · arxiv created 2021/06/19 · arxiv updated 2021/06/22 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28

Abstract

Actors and critics in actor-critic reinforcement learning algorithms are functionally separate, yet they often use the same network architectures. This case study explores the performance impact of network sizes when considering actor and critic architectures independently. By relaxing the assumption of architectural symmetry, it is often possible for smaller actors to achieve comparable policy performance to their symmetric counterparts. Our experiments show up to 99% reduction in the number of network weights with an average reduction of 77% over multiple actor-critic algorithms on 9 independent tasks. Given that reducing actor complexity results in a direct reduction of run-time inference cost, we believe configurations of actors and critics are aspects of actor-critic design that deserve to be considered independently, particularly in resource-constrained applications or when deploying multiple actors simultaneously.

Cited by

Related