Batch Reinforcement Learning with Hyperparameter Gradients.
Byung-Jun LeeJongmin LeePeter VrancxDongho KimKee-Eung KimPublished in: ICML (2020)
Keyphrases
- reinforcement learning
- batch mode
- function approximation
- markov decision processes
- reinforcement learning algorithms
- gaussian processes
- gaussian process
- robotic control
- machine learning
- model selection
- state space
- temporal difference
- learning algorithm
- quality prediction
- model free
- dynamic programming
- batch size
- autonomous learning
- optimal control
- optimal policy
- action selection
- regression model
- markov decision process
- control policy
- multi agent
- hyperparameters
- transition model
- policy search
- incremental learning