Skill-aware Mutual Information Optimisation for Generalisation in Reinforcement Learning.
Xuehui YuMhairi DunionXin LiStefano V. AlbrechtPublished in: CoRR (2024)
Keyphrases
- mutual information
- reinforcement learning
- information theoretic
- similarity measure
- feature selection
- function approximation
- image registration
- multi modal image registration
- medical image registration
- markov decision processes
- conditional mutual information
- reinforcement learning algorithms
- optimal policy
- temporal difference
- machine learning
- multi modality
- statistical dependence
- multi agent
- model free
- genetic algorithm
- kullback leibler divergence
- state space
- learning algorithm
- supervised learning
- reinforcement learning methods
- behavioural cloning
- robotic control
- information theoretic measures
- multi agent reinforcement learning
- mutual information maximization
- information theoretic criterion
- data mining
- skills needed
- normalized mutual information
- stochastic approximation
- action space
- information gain