Hyperparameter Tuning in Offline Reinforcement Learning.
Andrew TittaferranteAbdulsalam YassinePublished in: ICMLA (2022)
Keyphrases
- reinforcement learning
- state space
- real time
- function approximation
- gaussian processes
- reinforcement learning algorithms
- machine learning
- model selection
- markov decision processes
- parameter tuning
- multi agent
- learning problems
- transition model
- hyperparameters
- gaussian process
- temporal difference
- control policy
- markov decision process
- policy search
- fine tuning
- neural network
- model free
- parameter settings
- supervised learning
- dynamic programming
- learning algorithm
- rule selection