Litcius/Paper detail

Soft Contrastive Learning With Q-Irrelevance Abstraction for Reinforcement Learning

Minsong Liu, Luntong Li, Shuai Hao, Yuanheng Zhu, Dongbin Zhao

2022IEEE Transactions on Cognitive and Developmental Systems11 citationsDOI

Abstract

The difference between training and testing environments is a huge challenge to generalizing reinforcement learning (RL) algorithms. We propose a soft contrastive learning with a coarser approximate <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula> -irrelevance abstraction for RL (SCQRL) to increase RL generalization. Specifically, we specify the coarser approximate <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula> -irrelevance abstraction as the feature of the state with a theoretical analysis for better generalization ability. We construct a positive and negative sample selection mechanism based on the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula> value for contrastive learning to achieve efficient representation learning. Considering the selection error of positive and negative samples, we design soft contrastive learning and combine it with RL in the form of an auxiliary task to propose SCQRL. The generalization experiments on several Procgen environments demonstrate that SCQRL outperforms the excellent generalized RL algorithm.

Topics & Concepts

GeneralizationAbstractionNotationReinforcement learningComputer scienceArtificial intelligenceSelection (genetic algorithm)Representation (politics)AlgorithmTheoretical computer scienceMathematicsArithmeticLawMathematical analysisPolitical sciencePhilosophyEpistemologyPoliticsDomain Adaptation and Few-Shot LearningMachine Learning and ELMReinforcement Learning in Robotics
Soft Contrastive Learning With Q-Irrelevance Abstraction for Reinforcement Learning | Litcius