Learning generative models of natural images

Authors:
Jiann-Ming Wu;Zheng-Han Lin
Affiliations:
Department of Applied Mathematics, National Donghwa University, Hualien, Taiwan, ROC;Department of Applied Mathematics, National Donghwa University, Hualien, Taiwan, ROC
Venue:
Neural Networks
Year:
2002

Citing 8
Cited 3

Dimension reduction, generalized deformable models and the development of ocularity and orientation

Neural Networks
Self-organization using Potts models

Neural Networks
The joint development of orientation and ocular dominance: role of constraints

Neural Computation
Faithful representation of separable distributions

Neural Computation
A fast fixed-point algorithm for independent component analysis

Neural Computation
Emergence of Phase- and Shift-Invariant Features by Decomposition of Natural Images into Independent Feature Subspaces

Neural Computation
Fast and robust fixed-point algorithms for independent component analysis

IEEE Transactions on Neural Networks
Independent component analysis using Potts models

IEEE Transactions on Neural Networks

Blind separation of fetal electrocardiograms by annealed expectation maximization

Neurocomputing
Independent component analysis based on marginal density estimation using weighted Parzen windows

Neural Networks
Annealed Kullback-Leibler divergence minimization for generalized TSP, spot identification and gene sorting

Neurocomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This work proposes an unsupervised learning process for analysis of natural images. The derivation is based on a generative model, a stochastic coin-flip process directly operating on many disjoint multivariate Gaussian distributions. Following the maximal likelihood principle and using the Potts encoding, the goodness-of-fit of the generative model to tremendous patches randomly sampled from natural images is quantitatively expressed by an objective function subject to a set of constraints. By further combination of the objective function and the minimal wiring criterion, we achieve a mixed integer and linear programming. A hybrid of the mean field annealing and the gradient descent method is applied to the mathematical framework and produces three sets of interactive dynamics for the learning process. Numerical simulations show that the learning process is effective for extraction of orientation, localization and bandpass features and the generative model can make an ensemble of a sparse code for natural images.