www.uhasselt.be
DSpace

Document Server@UHasselt >
Research >
Research publications >

Please use this identifier to cite or link to this item: http://hdl.handle.net/1942/3469

Title: Unsupervised learning by examples: On-line versus off-line
Authors: VAN DEN BROECK, Christian
REIMANN, Peter
Issue Date: 1996
Publisher: AMERICAN PHYSICAL SOC
Citation: PHYSICAL REVIEW LETTERS, 76(12). p. 2188-2191
Abstract: We study both on-line and off-line unsupervised learning from p random patterns which are uniformly distributed on the N-sphere with the exception of a single symmetry breaking orientation B, along which they may be arbitrarily distributed. Supervised learning from the same kind of patterns is included as a special case. In the thermodynamic limit N --> infinity with alpha = p/N fixed we calculate the overlap R(alpha)= B . J/\J\\B\ between the unknown ''true'' B and the optimal ''Bayes'' hypothesis J with particular emphasis on the small and large cu asymptotics and the phenomenon of retarded learning. Finally, we identify a cost function whose minimum reproduces the off-line Bayes overlap.
Notes: VandenBroeck, C, LIMBURGS UNIV CENTRUM,B-3590 DIEPENBEEK,BELGIUM.
URI: http://hdl.handle.net/1942/3469
DOI: 10.1103/PhysRevLett.76.2188
ISI #: A1996TZ98400051
ISSN: 0031-9007
Type: Journal Contribution
Appears in Collections: Research publications

Files in This Item:

There are no files associated with this item.

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.