Semi-unsupervised Weighted Maximum-Likelihood Estimation of Joint Densities for the Co-training of Adaptive Activation Functions.
Ilaria CastelliEdmondo TrentinPublished in: PSL (2011)
Keyphrases
- maximum likelihood estimation
- co training
- semi supervised
- activation function
- supervised learning
- semi supervised learning
- probability distribution
- maximum likelihood
- unlabeled data
- em algorithm
- multi view
- neural network
- labeled data
- single view
- parameter estimation
- unsupervised learning
- feed forward
- expectation maximization
- text classification
- hidden layer
- back propagation
- artificial neural networks
- active learning
- neural nets
- density function
- multilayer perceptron
- training examples
- bayesian networks
- pairwise
- probability density function
- radial basis function
- named entities
- basis functions
- decision trees
- computer vision
- machine learning