Integration of Stochastic Models by Minimizing -Divergence

  title={Integration of Stochastic Models by Minimizing -Divergence},
  author={Shun-ichi Amari},
  journal={Neural Computation},
When there are a number of stochastic models in the form of probability distributions, one needs to integrate them. Mixtures of distributions are frequently used, but exponential mixtures also provide a good means of integration. This letter proposes a one-parameter family of integration, called -integration, which includes all of these well-known integrations. These are generalizations of various averages of numbers such as arithmetic, geometric, and harmonic averages. There are psychophysical… CONTINUE READING
Highly Influential
This paper has highly influenced 18 other papers. REVIEW HIGHLY INFLUENTIAL CITATIONS
Highly Cited
This paper has 108 citations. REVIEW CITATIONS

From This Paper

Topics from this paper.


Publications citing this paper.
Showing 1-10 of 59 extracted citations

Parameter Learning for Alpha Integration

Neural Computation • 2013
View 14 Excerpts
Highly Influenced

Parameter Estimation for -GMM Based on Maximum Likelihood Criterion

Neural Computation • 2009
View 6 Excerpts
Highly Influenced

On alpha-mean of Kullback-Leibler divergences for subspace selection

2011 Eighth International Conference on Fuzzy Systems and Knowledge Discovery (FSKD) • 2011
View 4 Excerpts
Highly Influenced

Alpha-integration of multiple evidence

2010 IEEE International Conference on Acoustics, Speech and Signal Processing • 2010
View 7 Excerpts
Highly Influenced

Learning α-integration with partially-labeled data

2010 IEEE International Conference on Acoustics, Speech and Signal Processing • 2010
View 7 Excerpts
Highly Influenced

The Dual Voronoi Diagrams with Respect to Representational Bregman Divergences

2009 Sixth International Symposium on Voronoi Diagrams • 2009
View 8 Excerpts
Highly Influenced

alpha-Gaussian mixture modelling for speaker recognition

Pattern Recognition Letters • 2009
View 5 Excerpts
Highly Influenced

Alpha-Pooling for Convolutional Neural Networks

ArXiv • 2018
View 4 Excerpts
Highly Influenced

109 Citations

Citations per Year
Semantic Scholar estimates that this publication has 109 citations based on the available data.

See our FAQ for additional information.


Publications referenced by this paper.
Showing 1-10 of 20 references

Training Products of Experts by Minimizing Contrastive Divergence

Neural Computation • 2002
View 4 Excerpts
Highly Influenced

Illumination-invariance of Plateau’s midgray

J. Heller
Journal of Mathematical Psychology, • 2006
View 2 Excerpts

Neural representation for perception of wide-field visual flow in MST: Bidirectional transparent motion and its illusory aftereffect

E. Hida, H. Ohno, N. Hashimoto, S. Amari, H. Saito

Divergence measures and message passing

Thomas Minka
View 2 Excerpts

Means of Positive Numbers and Matrices

SIAM J. Matrix Analysis Applications • 2005
View 1 Excerpt

Divergence Function, Duality, and Convex Analysis

Neural Computation • 2004
View 2 Excerpts

Similar Papers

Loading similar papers…