Improved Production of Competitive Learning Rules with an Additional Term for Vector Quantization

 

Abstract

In this work, a general framework for developing learning rules with an added term (perturbation term) is presented. Many learning rules commonly cited in the specialized literature can be derived from this general framework. This framework allows us to introduce some knowledge about vector quantization (as an optimization problem) in the distortion function in order to derive a new learning rule that uses that information to avoid certain local minima of the distortion function, leading to better performance than classical models. Computational experiments in image compression show that our proposed rule, derived from this general framework, can achieve better results than simple competitive learning and other models, with codebooks of less distortion. © Springer-Verlag Berlin Heidelberg 2007.

Cites

The following graph plots the number of cites received by this work from its publication, on a yearly basis.

Citation

Please, cite this work as:

[Cas+07] E. M. Casermeiro, D. López-Rodríguez, G. G. Marín, et al. “Improved Production of Competitive Learning Rules with an Additional Term for Vector Quantization”. In: Adaptive and Natural Computing Algorithms, 8th International Conference, ICANNGA 2007, Warsaw, Poland, April 11-14, 2007, Proceedings, Part I. Ed. by B. Beliczynski, A. Dzielinski, M. Iwanowski and B. Ribeiro. Vol. 4431. Lecture Notes in Computer Science PART 1. cited By 1; Conference of 8th International Conference on Adaptive and Natural Computing Algorithms, ICANNGA 2007 ; Conference Date: 11 April 2007 Through 14 April 2007; Conference Code:71057. Warsaw: Springer, 2007, pp. 461-469. DOI: 10.1007/978-3-540-71618-1_51. URL: https://doi.org/10.1007/978-3-540-71618-1_51.

BibTeX

@InProceedings{Casermeiro2007b,
     author = {Enrique Mérida Casermeiro and Domingo López-Rodríguez and Gloria Galán Marín and Juan Miguel Ortiz-de-Lazcano-Lobato},
     booktitle = {Adaptive and Natural Computing Algorithms, 8th International Conference, {ICANNGA} 2007, Warsaw, Poland, April 11-14, 2007, Proceedings, Part {I}},
     title = {Improved Production of Competitive Learning Rules with an Additional Term for Vector Quantization},
     year = {2007},
     address = {Warsaw},
     editor = {Bartlomiej Beliczynski and Andrzej Dzielinski and Marcin Iwanowski and Bernardete Ribeiro},
     note = {cited By 1; Conference of 8th International Conference on Adaptive and Natural Computing Algorithms, ICANNGA 2007 ; Conference Date: 11 April 2007 Through 14 April 2007; Conference Code:71057},
     number = {PART 1},
     pages = {461–469},
     publisher = {Springer},
     series = {Lecture Notes in Computer Science},
     volume = {4431},
     abstract = {In this work, a general framework for developing learning rules with an added term (perturbation term) is presented. Many learning rules commonly cited in the specialized literature can be derived from this general framework. This framework allows us to introduce some knowledge about vector quantization (as an optimization problem) in the distortion function in order to derive a new learning rule that uses that information to avoid certain local minima of the distortion function, leading to better performance than classical models. Computational experiments in image compression show that our proposed rule, derived from this general framework, can achieve better results than simple competitive learning and other models, with codebooks of less distortion. © Springer-Verlag Berlin Heidelberg 2007.},
     bibsource = {dblp computer science bibliography, https://dblp.org},
     biburl = {https://dblp.org/rec/conf/icannga/CasermeiroLMO07.bib},
     document_type = {Conference Paper},
     doi = {10.1007/978-3-540-71618-1_51},
     journal = {Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)},
     keywords = {Mathematical models; Perturbation techniques; Vector quantization, Competitive learning; Distortion function; Learning rules; Specialized literature, Learning systems},
     source = {Scopus},
     url = {https://doi.org/10.1007/978-3-540-71618-1_51},
}