Multi-Task Policy Search.
Marc Peter DeisenrothPeter EnglertJan PetersDieter FoxPublished in: CoRR (2013)
Keyphrases
- multi task
- policy search
- multi task learning
- reinforcement learning
- learning tasks
- continuous state
- reinforcement learning algorithms
- learning problems
- metric learning
- transfer learning
- feature selection
- dynamic programming
- multi class
- policy gradient
- state space
- finite state
- reward function
- function approximators
- partially observable markov decision processes
- learning algorithm
- decision trees
- learning process
- information gain
- neural network
- machine learning algorithms
- graphical models