Modeling-Learning-Based Actor-Critic Algorithm with Gaussian Process Approximator.
Shan ZhongJack TanHusheng DongXuemei ChenShengrong GongZhenjiang QianPublished in: J. Grid Comput. (2020)
Keyphrases
- actor critic
- learning algorithm
- gaussian process
- reinforcement learning
- dynamic programming
- objective function
- optimal solution
- expectation maximization
- regression model
- policy gradient
- reinforcement learning algorithms
- optimal control
- bayesian framework
- monte carlo
- active learning
- markov decision processes
- np hard
- learning process
- search space
- approximate dynamic programming
- machine learning
- model selection
- supervised learning
- state space
- convergence rate
- incremental learning
- prior knowledge
- gaussian processes
- temporal difference