Gaussian-binary Restricted Boltzmann Machines on Modeling Natural Image Statistics

Abstract

We present a theoretical analysis of Gaussian-binary restricted Boltzmann machines (GRBMs) from the perspective of density models. The key aspect of this analysis is to show that GRBMs can be formulated as a constrained mixture of Gaussians, which gives a much better insight into the model's capabilities and limitations. We show that GRBMs are capable of learning meaningful features both in a two-dimensional blind source separation task and in modeling natural images. Further, we show that reported difficulties in training GRBMs are due to the failure of the training algorithm rather than the model itself. Based on our analysis we are able to propose several training recipes, which allowed successful and fast training in our experiments. Finally, we discuss the relationship of GRBMs to several modifications that have been proposed to improve the model.

Current version is only an early manuscript and is subject to further change

References in corpus (3)

• Iterative Gaussianization: from ICA to Random Rotations
• Gaussian-binary Restricted Boltzmann Machines on Modeling Natural Image Statistics
• How to Center Binary Deep Boltzmann Machines

Cited by in corpus (8)

• Gaussian-binary Restricted Boltzmann Machines on Modeling Natural Image Statistics
• A Deterministic and Generalized Framework for Unsupervised Learning with Restricted Boltzmann Machines
• Riemann-Theta Boltzmann Machine
• On the mapping between Hopfield networks and Restricted Boltzmann Machines
• Restricted Boltzmann Machines with Gaussian Visible Units Guided by Pairwise Constraints
• Efficient Tensor Decomposition with Boolean Factors
• Shared latent subspace modelling within Gaussian-Binary Restricted Boltzmann Machines for NIST i-Vector Challenge 2014
• Learning Gaussian-Bernoulli RBMs using Difference of Convex Functions Optimization