Document Detail


Window-based example selection in learning vector quantization.
MedLine Citation:
PMID:  20804387     Owner:  NLM     Status:  In-Process    
Abstract/OtherAbstract:
A variety of modifications have been employed to learning vector quantization (LVQ) algorithms using either crisp or soft windows for selection of data. Although these schemes have been shown in practice to improve performance, a theoretical study on the influence of windows has so far been limited. Here we rigorously analyze the influence of windows in a controlled environment of gaussian mixtures in high dimensions. Concepts from statistical physics and the theory of online learning allow an exact description of the training dynamics, yielding typical learning curves, convergence properties, and achievable generalization abilities. We compare the performance and demonstrate the advantages of various algorithms, including LVQ 2.1, generalized LVQ (GLVQ), Learning from Mistakes (LFM) and Robust Soft LVQ (RSLVQ). We find that the selection of the window parameter highly influences the learning curves but not, surprisingly, the asymptotic performances of LVQ 2.1 and RSLVQ. Although the prototypes of LVQ 2.1 exhibit divergent behavior, the resulting decision boundary coincides with the optimal decision boundary, thus yielding optimal generalization ability.
Authors:
A W Witoelar; A Ghosh; J J G de Vries; B Hammer; M Biehl
Related Documents :
19497817 - Softdoublemaxminover: perceptron-like training of support vector machines.
14618827 - Use of a program logic model to guide the development of a strategic plan for wellingto...
18255767 - Evolution and development of neural controllers for locomotion, gradient-following, and...
18423077 - Thrips (thysanoptera) identification using artificial neural networks.
21986697 - Physical fitness of elite belgian soccer players by player position.
18550387 - The neural basis of academic achievement motivation.
Publication Detail:
Type:  Journal Article    
Journal Detail:
Title:  Neural computation     Volume:  22     ISSN:  1530-888X     ISO Abbreviation:  Neural Comput     Publication Date:  2010 Nov 
Date Detail:
Created Date:  2010-10-13     Completed Date:  -     Revised Date:  -    
Medline Journal Info:
Nlm Unique ID:  9426182     Medline TA:  Neural Comput     Country:  United States    
Other Details:
Languages:  eng     Pagination:  2924-61     Citation Subset:  IM    
Affiliation:
Johann Bernoulli Institute for Mathematics and Computer Science, University of Groningen, Groningen, Netherlands. a.w.witoelar@rug.nl
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine


Previous Document:  Comparing classification methods for longitudinal fMRI studies.
Next Document:  The regulation of differentiation in mesenchymal stem cells.