Keyword : information geometry


Boltzmann Machines with Identified States
Masaki KOBAYASHI 
Publication:   IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences
Publication Date: 2008/03/01
Vol. E91-A  No. 3 ; pp. 887-890
Type of Manuscript:  LETTER
Category: Nonlinear Problems
Keyword: 
boltzmann machineneural networkboltzmann learninginformation geometryem algorithm
 Summary | Full Text:PDF(71.4KB)

Geometric Properties of Quasi-Additive Learning Algorithms
Kazushi IKEDA 
Publication:   IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences
Publication Date: 2006/10/01
Vol. E89-A  No. 10 ; pp. 2812-2817
Type of Manuscript:  Special Section PAPER (Special Section on Nonlinear Theory and its Applications)
Category: Control, Neural Networks and Learning
Keyword: 
quasi-additive algorithmsperceptroninformation geometry
 Summary | Full Text:PDF(208KB)

Independent Component Analysis (ICA) and Method of Estimating Functions
Shun-ichi AMARI 
Publication:   IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences
Publication Date: 2002/03/01
Vol. E85-A  No. 3 ; pp. 540-547
Type of Manuscript:  INVITED PAPER (Special Section on the Trend of Digital Signal Processing and Its Future Direction)
Category: Theories
Keyword: 
independent componentsinformation geometryestimating functionslearningsignal processing
 Summary | Full Text:PDF(204.3KB)

A New Relation between Information Geometry and Convex Programming--Coincidence with the Gradient Vectors for the Divergence and a Modified Barrier Function
Yukio HAYASHI 
Publication:   IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences
Publication Date: 2001/09/01
Vol. E84-A  No. 9 ; pp. 2238-2246
Type of Manuscript:  Special Section PAPER (Special Section on Nonlinear Theory and its Applications)
Category: Numerical Method & Optimization
Keyword: 
information geometrygradient systeminterior point methodself-concordant barrier functionlp-programming
 Summary | Full Text:PDF(330.3KB)

Differential and Algebraic Geometry of Multilayer Perceptrons
Shun-ichi AMARI Tomoko OZEKI 
Publication:   IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences
Publication Date: 2001/01/01
Vol. E84-A  No. 1 ; pp. 31-38
Type of Manuscript:  INVITED PAPER (Special Section on the 10th Anniversary of the IEICE Transactions of Fundamentals: "Last Decade and 21st Century")
Category: 
Keyword: 
information geometrymultilayer perceptronsingularitieslearningnatural gradient
 Summary | Full Text:PDF(385.7KB)

Divergence-Based Geometric Clustering and Its Underlying Discrete Proximity Structures
Hiroshi IMAI Mary INABA 
Publication:   IEICE TRANSACTIONS on Information and Systems
Publication Date: 2000/01/25
Vol. E83-D  No. 1 ; pp. 27-35
Type of Manuscript:  INVITED PAPER (Special Issue on Surveys on Discovery Science)
Category: 
Keyword: 
unsupervised learninggeometric clusteringVoronoi diagramcomputational geometryinformation geometry
 Summary | Full Text:PDF(349.8KB)

Direct Calculation Methods for Parameter Estimation in Statistical Manifolds of Finite Discrete Distributions
Yukio HAYASHI 
Publication:   IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences
Publication Date: 1998/07/25
Vol. E81-A  No. 7 ; pp. 1486-1492
Type of Manuscript:  PAPER
Category: General Fundamentals and Boundaries
Keyword: 
information geometryparameter estimationfinite discrete distributionexponential family mixture familyautoparallel submanifoldgradient system
 Summary | Full Text:PDF(513KB)

On the Twisted Markov Chain of Importance Sampling Simulation
Kenji NAKAGAWA 
Publication:   IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences
Publication Date: 1996/09/25
Vol. E79-A  No. 9 ; pp. 1423-1428
Type of Manuscript:  Special Section PAPER (Special Section on Information Theory and Its Applications)
Category: Stochastic Process/Learning
Keyword: 
importance samplingMarkov chainKullback-Leibler divergencelarge deviations theoryinformation geometry
 Summary | Full Text:PDF(437.3KB)

Information Geometry of Mean Field Theory
Toshiyuki TANAKA 
Publication:   IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences
Publication Date: 1996/05/25
Vol. E79-A  No. 5 ; pp. 709-715
Type of Manuscript:  PAPER
Category: Neural Networks
Keyword: 
neural networksmean field theorysimulated annealinginformation geometryoptimization problem.
 Summary | Full Text:PDF(547.5KB)

Information Geometry of Neural Networks
Shun-ichi AMARI 
Publication:   IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences
Publication Date: 1992/05/25
Vol. E75-A  No. 5 ; pp. 531-536
Type of Manuscript:  INVITED PAPER (Special Section on Nonlinear Dynamics--Adaptive, Learning and Neural Systems--)
Category: 
Keyword: 
neural networksinformation geometrynon-linear systemprobability distribution
 Summary | Full Text:PDF(471.9KB)