Policy Optimization with Augmented Value Targets for Generalization in Reinforcement Learning.
Nasik Muhammad NafiGiovanni Poggi-CorradiniWilliam H. HsuPublished in: IJCNN (2023)
Keyphrases
- reinforcement learning
- optimal policy
- markov decision process
- policy search
- function approximation
- optimization algorithm
- action selection
- machine learning
- reinforcement learning problems
- markov decision processes
- optimization method
- partially observable domains
- state and action spaces
- control policy
- reinforcement learning algorithms
- optimization methods
- state space
- global optimization
- finite state
- neural network
- learning algorithm
- control policies
- optimization problems
- partially observable
- constrained optimization
- policy iteration
- supervised learning
- approximate dynamic programming
- reward function
- dynamic programming
- temporal difference
- partially observable environments