Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/mit-press/soft-learning-vector-quantization-wqPlLkNNxT
References (11)
-
T. Kohonen (1990)
Improved versions of learning vector quantization1990 IJCNN International Joint Conference on Neural Networks
-
(2002)
Bibliography on the self-organizing map (SOM) and learning vector quantization (LVQ) -
T. Kohonen (1998)
Learning vector quantization -
T. Komori, S. Katagiri (1992)
Application of a generalized probabilistic descent method to dynamic time warping-based speech recognition[Proceedings] ICASSP-92: 1992 IEEE International Conference on Acoustics, Speech, and Signal Processing, 1
-
E. McDermott, S. Katagiri (1994)
Prototype-based minimum classification error/generalized probabilistic descent training for various speech unitsComput. Speech Lang., 8
-
T. Graepel, M. Burger, K. Obermayer (1997)
PHASE TRANSITIONS IN STOCHASTIC SELF-ORGANIZING MAPSPhysical Review E, 56
-
T. Kohonen, J. Hynninen, J. Kangas, Jorma Laaksonen, K. Torkkola (1996)
LVQ_PAK: The Learning Vector Quantization Program Package -
R. Duda, P. Hart (1974)
Pattern classification and scene analysis -
David Miller, A. Rao, K. Rose, A. Gersho (1996)
A global optimization technique for statistical classifier designIEEE Trans. Signal Process., 44
-
J. Suykens, J. Vandewalle (1999)
Multiclass least squares support vector machinesIJCNN'99. International Joint Conference on Neural Networks. Proceedings (Cat. No.99CH36339), 2
-
(2001)
Self organization maps
- Publisher
- MIT Press
- Copyright
- © 2003 Massachusetts Institute of Technology
- ISSN
- 0899-7667
- eISSN
- 1530-888X
- DOI
- 10.1162/089976603321891819
- pmid
- 12816567
- Publisher site
- See Article on Publisher Site
Abstract
Learning vector quantization (LVQ) is a popular class of adaptive nearest prototype classifiers for multiclass classification, but learning algorithms from this family have so far been proposed on heuristic grounds. Here, we take a more principled approach and derive two variants of LVQ using a gaussian mixture ansatz. We propose an objective function based on a likelihood ratio and derive a learning rule using gradient descent. The new approach provides a way to extend the algorithms of the LVQ family to different distance measure and allows for the design of “soft” LVQ algorithms. Benchmark results show that the new methods lead to better classification performance than LVQ 2.1. An additional benefit of the new method is that model assumptions are made explicit, so that the method can be adapted more easily to different kinds of problems.
Journal
Neural Computation – MIT Press
Published: Jul 1, 2003