Self-Supervised Reinforcement Learning for Out-of-Distribution Recovery via Auxiliary Reward.
Yufeng XieYinan WangHan WangQingshan LiPublished in: ICASSP (2024)
Keyphrases
- reinforcement learning
- function approximation
- eligibility traces
- state space
- reinforcement learning algorithms
- markov decision processes
- learning algorithm
- neural network
- spatial distribution
- optimal control
- mobile robot
- machine learning
- optimal policy
- partially observable environments
- data sets
- function approximators
- learning capabilities
- multi agent
- gaussian distribution
- action selection
- temporal difference
- reward function
- uniformly distributed
- partially observable
- error detection
- probability distribution
- state action
- initially unknown
- least squares
- supervised learning