Global optimality of softmax policy gradient with single hidden layer neural networks in the mean-field regime.
Andrea AgazziJianfeng LuPublished in: CoRR (2020)
Keyphrases
- activation function
- neural network
- global optimality
- hidden layer
- artificial neural networks
- back propagation
- feed forward
- global optimization
- hidden nodes
- multilayer perceptron
- optimal solution
- learning rate
- basis functions
- radial basis function
- global minimum
- markov random field
- em algorithm
- cost function
- markov networks
- reinforcement learning
- upper bound
- rbf neural network
- fuzzy neural network
- bayesian networks
- function approximation
- recurrent neural networks
- search space