Publication: BarlowRL: barlow twins for data-efficient reinforcement learning
Program
KU-Authors
KU Authors
Co-Authors
Publication Date
Language
Embargo Status
Journal Title
Journal ISSN
Volume Title
Alternative Title
Abstract
This paper introduces BarlowRL, a data-efficient reinforcement learning agent that combines the Barlow Twins self-supervised learning framework with DER (Data-Efficient Rainbow) algorithm. BarlowRL outperforms both DER and its contrastive counterpart CURL on the Atari 100k benchmark. BarlowRL avoids dimensional collapse by enforcing information spread to the whole space. This helps RL algorithms to utilize uniformly spread state representation that eventually results in a remarkable performance. The integration of Barlow Twins with DER enhances data efficiency and achieves superior performance in the RL tasks. BarlowRL demonstrates the potential of incorporating self-supervised learning techniques, especially that of non-contrastive objectives, to improve RL algorithms.
Source
Publisher
JMLR-Jornal Machine Learning Research
Subject
Computer science, artificial intelligence, Computer science, theory and methods, Statistics and probability
Citation
Has Part
Source
Asian Conference on Machine Learning Vol 222