Bootstrap Advantage Estimation for Policy Optimization in Reinforcement Learning.
Md. Masudur RahmanYexiang XuePublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- optimal policy
- policy search
- action selection
- function approximation
- markov decision process
- function approximators
- optimization problems
- markov decision processes
- optimization algorithm
- optimization method
- action space
- partially observable
- global optimization
- partially observable environments
- cross validation
- optimization process
- estimation algorithm
- state and action spaces
- multi agent
- state dependent
- reinforcement learning problems
- actor critic
- approximate dynamic programming
- policy gradient
- partially observable domains
- long run
- policy iteration
- reinforcement learning algorithms
- temporal difference
- model free
- neural network
- state space
- dynamic programming
- evolutionary algorithm
- learning algorithm
- machine learning