Deep Boltzmann Machines

@inproceedings{Salakhutdinov2009DeepBM,
  title={Deep Boltzmann Machines},
  author={Ruslan Salakhutdinov and Geoffrey E. Hinton},
  booktitle={AISTATS},
  year={2009}
}
We present a new learning algorithm for Boltzmann machines that contain many layers of hidden variables. Data-dependent expectations are estimated using a variational approximation that tends to focus on a single mode, and dataindependent expectations are approximated using persistent Markov chains. The use of two quite different techniques for estimating the two types of expectation that enter into the gradient of the log-likelihood makes it practical to learn Boltzmann machines with multiple… CONTINUE READING

Similar Papers

Citations

Publications citing this paper.
SHOWING 1-10 OF 1,094 CITATIONS, ESTIMATED 88% COVERAGE

Deep Recurrent Learning for Efficient Image Recognition Using Small Data

VIEW 25 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

Exponential Family Restricted Boltzmann Machines and Annealed Importance Sampling

  • 2018 International Joint Conference on Neural Networks (IJCNN)
  • 2018
VIEW 7 EXCERPTS
CITES BACKGROUND
HIGHLY INFLUENCED

Deep Learning the Ising Model Near Criticality

  • J. Mach. Learn. Res.
  • 2017
VIEW 5 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

GSNs : Generative Stochastic Networks

VIEW 11 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Deep learning of representations and its application to computer vision

VIEW 17 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

FILTER CITATIONS BY YEAR

2009
2019

CITATION STATISTICS

  • 201 Highly Influenced Citations

  • Averaged 149 Citations per year over the last 3 years