A bound for the convergence rate of parallel tempering for sampling restricted Boltzmann machines.
Asja FischerChristian IgelPublished in: Theor. Comput. Sci. (2015)
Keyphrases
- restricted boltzmann machine
- convergence rate
- deep learning
- learning rate
- convergence speed
- hidden layer
- probabilistic graphical models
- conditional random fields
- semi supervised
- learning algorithm
- random sampling
- loopy belief propagation
- semi supervised learning
- graphical models
- markov chain monte carlo
- belief propagation
- radial basis function
- genetic algorithm
- markov random field
- approximate inference
- least squares
- support vector