Preprint
SimSR: Simple Distance-based State Representation for Deep Reinforcement Learning
arXiv, Vol.27 February 2022, 2112.15303v2
Cornell University
2021
Abstract
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, which achieves equivalent functionality while reducing the complexity by an order in comparison with bisimulation metric. SimSR enables us to design a stochastic-approximation-based method that can practically learn the mapping functions (encoders) from observations to latent representation space. Besides the theoretical analysis, 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 Representation for Deep Reinforcement Learning
- Authors
- Hongyu Zang (Author) - Beijing Institute of TechnologyXin Li (Author) - Beijing Institute of TechnologyMingzhong Wang (Author) - University of the Sunshine Coast, Queensland, School of Science, Technology and Engineering
- Publication details
- arXiv, Vol.27 February 2022, 2112.15303v2
- Publisher
- Cornell University
- DOI
- 10.48550/arXiv.2112.15303
- ISSN
- 2331-8422
- Organisation Unit
- University of the Sunshine Coast, Queensland; School of Science, Technology and Engineering
- Language
- English
- Record Identifier
- 99605107202621
- Output Type
- Preprint
Metrics
6 Record Views