A Decentralized Policy Gradient Approach to Multi-task Reinforcement Learning.
Sihan ZengAqeel AnwarThinh T. DoanJustin RombergArijit RaychowdhuryPublished in: CoRR (2020)
Keyphrases
- multi task
- policy gradient
- reinforcement learning
- actor critic
- transfer learning
- learning tasks
- function approximation
- learning problems
- reinforcement learning algorithms
- multi agent
- gradient method
- optimal control
- reinforcement learning methods
- single agent
- metric learning
- multi class
- learning algorithm
- approximation methods
- feature selection
- state space
- machine learning
- model free
- function approximators
- state action
- supervised learning
- rl algorithms
- temporal difference
- markov decision processes
- dynamic programming
- training data
- machine learning algorithms