Federated Natural Policy Gradient Methods for Multi-task Reinforcement Learning.
Tong YangShicong CenYuting WeiYuxin ChenYuejie ChiPublished in: CoRR (2023)
Keyphrases
- multi task
- policy gradient methods
- reinforcement learning
- multi task learning
- natural actor critic
- transfer learning
- learning problems
- learning tasks
- policy gradient
- multi class
- actor critic
- feature selection
- function approximators
- function approximation
- robot arm
- learning algorithm
- state space
- metric learning
- reinforcement learning methods
- reinforcement learning algorithms
- machine learning
- markov decision processes
- single agent
- learning process
- dynamic programming
- machine learning algorithms
- semi supervised learning
- supervised learning