Hyp-RL : Hyperparameter Optimization by Reinforcement Learning.
Hadi S. JomaaJosif GrabockaLars Schmidt-ThiemePublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- function approximation
- state space
- reinforcement learning algorithms
- model free
- control problems
- optimization problems
- learning algorithm
- markov decision processes
- continuous state
- model selection
- reinforcement learning methods
- transfer learning
- optimal policy
- supervised learning
- learning problems
- autonomous learning
- optimal control
- gaussian processes
- temporal difference
- multi agent reinforcement learning
- hyperparameters
- action selection
- gaussian process
- optimization algorithm
- learning process
- learning classifier systems
- learning capabilities
- closed form
- markov decision process
- maximum likelihood
- active learning
- exploration exploitation
- direct policy search