SimSR: Simple Distance-Based State Representations for Deep Reinforcement Learning

Hongyu Zang; Xin Li; Mingzhong Wang

doi:10.1609/aaai.v36i8.20883

Back

SimSR: Simple Distance-Based State Representations for Deep Reinforcement Learning

Conference paper

Peer reviewed

SimSR: Simple Distance-Based State Representations for Deep Reinforcement Learning

Hongyu Zang, Xin Li and Mingzhong Wang

Proceedings of the AAAI Conference on Artificial Intelligence, Vol.36(8), pp.8997-9005

Association for the Advancement of Artificial Intelligence (AAAI) Conference on Artificial Intelligence, 36th (Online, 22-Feb-2022–01-Mar-2022)

Association for the Advancement of Artificial Intelligence

2022

DOI: https://doi.org/10.1609/aaai.v36i8.20883

Files and links (2)

url

https://doi.org/10.1609/aaai.v36i8.20883View

Published Version

url

https://aaai.org/Conferences/AAAI-22/View

Event Website

Abstract

Machine learning

machine learning (ML)

This work explores how to learn robust and generalizable state representation from image-based observations with deep reinforcement learning methods. Addressing the computational complexity, stringent assumptions and representation collapse challenges in existing work of bisimulation metric, we devise Simple State Representation (SimSR) operator. SimSR enables us to design a stochastic approximation method that can practically learn the mapping functions (encoders) from observations to latent representation space. In addition to the theoretical analysis and comparison with the existing work, we experimented and compared our work with recent state-of-the-art solutions in visual MuJoCo tasks. The results shows that our model generally achieves better performance and has better robustness and good generalization.

Details

Title: SimSR: Simple Distance-Based State Representations for Deep Reinforcement Learning
Authors: Hongyu Zang (Author) - Beijing Institute of Technology
Xin Li (Corresponding Author) - Beijing Institute of Technology
Mingzhong Wang (Author) - University of the Sunshine Coast, Queensland, School of Science, Technology and Engineering
Publication details: Proceedings of the AAAI Conference on Artificial Intelligence, Vol.36(8), pp.8997-9005
Conference details: Association for the Advancement of Artificial Intelligence (AAAI) Conference on Artificial Intelligence, 36th (Online, 22-Feb-2022–01-Mar-2022)
Publisher: Association for the Advancement of Artificial Intelligence
Date published: 2022
DOI: 10.1609/aaai.v36i8.20883
Organisation Unit: School of Science, Technology and Engineering; University of the Sunshine Coast, Queensland
Language: English
Record Identifier: 99689798402621
Output Type: Conference paper

Metrics

63 Record Views