Summary: Information-Theoretic Clustering in Nonlinear
Encoder Models
Felix Agakov
(University of Edinburgh, UK)
David Barber
(IDIAP, Switzerland)
July 4, 2005
Information-Theoretic Clustering in Nonlinear Encoder Models ­ p. 1/1
Overview
Information-theoretic clustering in encoder models:
conceptually simple
probabilistic (soft)
kernelizable and applicable to unsupervized learning of
kernel functions
computationally attractive (no need to compute
eigenvalues or inverses of the Gram matrix)
favorably compares with common clustering methods in
some cases
Information-Theoretic Clustering in Nonlinear Encoder Models ­ p. 2/1
Overview

Source: Agakov, Felix - Institute for Adaptive and Neural Computation, School of Informatics, University of Edinburgh

Collections: Computer Technologies and Information Sciences