An Adaptive Learning Method of Restricted Boltzmann Machine by Neuron Generation and Annihilation Algorithm.
Shin KamadaTakumi IchimuraPublished in: CoRR (2018)
Keyphrases
- learning algorithm
- dynamic programming
- support vector machine svm
- k means
- segmentation method
- similarity measure
- convergence rate
- prior knowledge
- energy function
- segmentation algorithm
- learning process
- probabilistic model
- parameter estimation
- unsupervised learning
- markov random field
- em algorithm
- expectation maximization
- higher order
- bayesian framework
- graph cuts
- maximum likelihood
- neural network
- support vector machine
- active learning
- reinforcement learning
- image segmentation
- markov chain monte carlo