An adaptive learning method of Restricted Boltzmann Machine by neuron generation and annihilation algorithm.
Shin KamadaTakumi IchimuraPublished in: SMC (2016)
Keyphrases
- learning algorithm
- dynamic programming
- k means
- segmentation algorithm
- similarity measure
- energy function
- support vector machine svm
- convergence rate
- prior knowledge
- segmentation method
- bayesian framework
- supervised learning
- parameter estimation
- restricted boltzmann machine
- neural network
- unsupervised learning
- em algorithm
- reinforcement learning
- probabilistic model
- learning process
- neural nets
- probabilistic graphical models