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/1942/7803

Title: On phase transitions in learning sparse networks
Authors: HOLLANDERS, Goele
BEX, Geert Jan
GYSSENS, Marc
WESTRA, Ronald
TUYLS, Karl
Issue Date: 2007
Publisher: SPRINGER-VERLAG BERLIN
Citation: MACHINE LEARNING: ECML 2007, PROCEEDINGS, 4701. p. 591-599
Series/Report: LECTURE NOTES IN COMPUTER SCIENCE, 4701
Abstract: In this paper we study the identification of sparse interaction networks as a machine learning problem. Sparsity means that we are provided with a small data set and a high number of unknown components of the system, most of which are zero. Under these circumstances, a model needs to be learned that fits the underlying system, capable of generalization. This corresponds to the student-teacher setting in machine learning. In the first part of this paper we introduce a learning algorithm, based on L-1-minimization, to identify interaction networks from poor data and analyze its dynamics with respect to phase transitions. The efficiency of the algorithm is measured by the generalization error, which represents the probability that the student is a good fit to the teacher. In the second part of this paper we show that from a system with a specific system size value the generalization error of other system sizes can be estimated. A comparison with a set of simulation experiments show a very good fit.
Notes: Hasselt Univ, Dept Math Phys & Comp Sci, Hasselt, Belgium.Hollanders, G, Hasselt Univ, Dept Math Phys & Comp Sci, Hasselt, Belgium.
URI: http://hdl.handle.net/1942/7803
DOI: 10.1007/978-3-540-74958-5_57
ISI #: 000249742300053
ISBN: 978-3-540-74957-8
ISSN: 0302-9743
Category: A1
Type: Journal Contribution
Validation: ecoom, 2008
Appears in Collections: Research publications

Files in This Item:

Description SizeFormat
N/A192.51 kBAdobe PDF

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