Training Restricted Boltzmann Machines Using Modified Objective Function Based on Limiting the Free Energy Value.
Song GuoChangjun ZhouBin WangXuedong ZhengPublished in: IEEE Access (2018)
Keyphrases
- restricted boltzmann machine
- free energy
- loopy belief propagation
- contrastive divergence
- objective function
- belief propagation
- deep learning
- learning algorithm
- fixed point
- probabilistic graphical models
- upper bound
- competitive learning
- conditional random fields
- posterior distribution
- semi supervised
- approximate inference
- optical flow
- graphical models
- least squares
- optimal solution
- lower bound
- training set
- message passing
- markov random field
- bayesian networks
- machine learning