Learning a subspace of policies for online adaptation in Reinforcement Learning.
Jean-Baptiste GayaLaure SoulierLudovic DenoyerPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- online learning
- learning algorithm
- learning problems
- learning process
- learning systems
- supervised learning
- dynamic programming
- optimal policy
- actor critic
- learning tasks
- learning capabilities
- learning agent
- policy search
- online environment
- neural network
- temporal difference learning
- learning agents
- subspace learning
- optimal control
- markov decision processes
- transfer learning
- feature extraction