Measuring the usefulness of hidden units in Boltzmann machines with mutual information.
Mathias BerglundTapani RaikoKyungHyun ChoPublished in: Neural Networks (2015)
Keyphrases
- boltzmann machine
- boltzmann machines
- hidden units
- rbf network
- neural network
- feed forward neural networks
- hidden layer
- radial basis function
- nonlinear functions
- generative model
- maximum likelihood estimation
- global constraints
- learning tasks
- similarity measure
- activation function
- learning rules
- input space
- feed forward
- rbf neural network
- artificial neural networks
- learning algorithm
- multi layer
- learning rate
- kernel function
- image processing
- machine learning