Convergence of Actor-Critic with Multi-Layer Neural Networks.
Haoxing TianAlex OlshevskyYannis PaschalidisPublished in: NeurIPS (2023)
Keyphrases
- multi layer
- convergence proof
- actor critic
- neural network
- reinforcement learning
- temporal difference
- gradient method
- neural nets
- policy gradient
- feed forward neural networks
- optimal control
- neuro fuzzy
- approximate dynamic programming
- reinforcement learning algorithms
- convergence rate
- function approximation
- markov decision processes
- policy iteration
- convergence speed
- learning tasks
- artificial neural networks
- semi supervised
- basis functions
- model free
- step size
- linear program