Ensemble of Independent Factor Analyzers with Application to Natural Image Analysis

Authors:
Akio Utsugi
Affiliations:
National Institute of Advanced Industrial Science and Technology, 1-1-1 Higashi Tsukuba Ibaraki 305-8566, Japan. E-mail: A28532@aist.go.jp
Venue:
Neural Processing Letters
Year:
2001

Citing 10
Cited 1

An information-maximization approach to blind separation and blind deconvolution

Neural Computation
Self-organizing maps

Self-organizing maps
Factorial Hidden Markov Models

Machine Learning - Special issue on learning with probabilistic representations
EM algorithms for PCA and SPCA

NIPS '97 Proceedings of the 1997 conference on Advances in neural information processing systems 10
Mixtures of probabilistic principal component analyzers

Neural Computation
Independent factor analysis

Neural Computation
Bayesian Analysis of Mixtures of Factor Analyzers

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

Neural Computation
Learning Overcomplete Representations

Neural Computation
Modeling the manifolds of images of handwritten digits

IEEE Transactions on Neural Networks

Nonlinear Boosting Projections for Ensemble Construction

The Journal of Machine Learning Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper the ensemble of independent factor analyzers (EIFA) is proposed. This new statistical model assumes that each data point is generated by the sum of outputs of independently activated factor analyzers. A maximum likelihood (ML) estimation algorithm for the parameter is derived using a Monte Carlo EM algorithm with a Gibbs sampler. The EIFA model is applied to natural image data. With the progress of the learning, the independent factor analyzers develop into feature detectors that resemble complex cells in mammalian visual systems. Although this result is similar to the previous one obtained by independent subspace analysis, we observe the emergence of complex cells from natural images in a more general framework of models, including overcomplete models allowing additive noise in the observables.