Stochastic Gradient Langevin dynamics that Exploit Neural Network Structure.
Zachary NadoJasper SnoekRoger B. GrosseDavid DuvenaudBowen XuJames MartensPublished in: ICLR (Workshop) (2018)
Keyphrases
- stochastic gradient
- neural network structure
- neural network
- nearest neighbor classifier
- training process
- multi layer perceptron
- stochastic gradient descent
- recurrent neural networks
- hidden layer
- fuzzy neural network
- training algorithm
- step size
- training data
- small number
- distance measure
- artificial neural networks
- feature space
- learning algorithm