Neural Network Approximation for Pessimistic Offline Reinforcement Learning.
Di WuYuling JiaoLi ShenHaizhao YangXiliang LuPublished in: CoRR (2023)
Keyphrases
- neural network
- reinforcement learning
- function approximators
- learning algorithm
- machine learning
- function approximation
- optimal policy
- back propagation
- approximation methods
- reinforcement learning algorithms
- neural nets
- fuzzy logic
- state space
- real time
- learning capabilities
- fault diagnosis
- self organizing maps
- transfer learning
- fuzzy artmap
- fermentation process
- feed forward
- neural network model
- radial basis function
- expert systems
- markov decision processes
- decision making
- error bounds
- multi agent
- artificial neural networks
- approximation error
- feed forward neural networks
- learning process
- knn
- expected utility
- action selection
- activation function
- network architecture
- hidden layer
- closed form
- approximation algorithms
- multilayer perceptron