Multitask reinforcement learning on the distribution of MDPs.
Fumihide TanakaMasayuki YamamuraPublished in: CIRA (2003)
Keyphrases
- reinforcement learning
- multi task
- markov decision processes
- transfer learning
- learning problems
- multi task learning
- multitask learning
- state space
- optimal policy
- function approximation
- learning tasks
- continuous state and action spaces
- markov decision process
- feature selection
- policy search
- gaussian processes
- dynamic programming
- learning algorithm
- model free
- reinforcement learning algorithms
- multiple tasks
- pairwise
- multiclass support vector machines
- supervised learning
- temporal difference
- policy iteration
- average reward
- learning styles
- reward function
- learning process
- continuous state spaces