Extended evaluation of the Volterra-Neural Network for model compression
When large models are used for a classification task, model compression is necessary because there are transmission, space, time or computing constraints that have to be fulfilled. Multilayer Perceptron (MLP) models are traditionally used as a classifier, but depending on the problem, they may need...
Guardado en:
Autor principal: | |
---|---|
Formato: | Objeto de conferencia |
Lenguaje: | Inglés |
Publicado: |
2012
|
Materias: | |
Acceso en línea: | http://sedici.unlp.edu.ar/handle/10915/123741 https://41jaiio.sadio.org.ar/sites/default/files/14_ASAI_2012.pdf |
Aporte de: |
id |
I19-R120-10915-123741 |
---|---|
record_format |
dspace |
institution |
Universidad Nacional de La Plata |
institution_str |
I-19 |
repository_str |
R-120 |
collection |
SEDICI (UNLP) |
language |
Inglés |
topic |
Ciencias Informáticas Model compression Volterra-Neural Network Extended evaluation |
spellingShingle |
Ciencias Informáticas Model compression Volterra-Neural Network Extended evaluation Rubiolo, Mariano Extended evaluation of the Volterra-Neural Network for model compression |
topic_facet |
Ciencias Informáticas Model compression Volterra-Neural Network Extended evaluation |
description |
When large models are used for a classification task, model compression is necessary because there are transmission, space, time or computing constraints that have to be fulfilled. Multilayer Perceptron (MLP) models are traditionally used as a classifier, but depending on the problem, they may need a large number of parameters (neuron functions, weights and bias) to obtain an acceptable performance. This work extends the evaluation of a technique to compress an array of MLPs, through the outputs of a Volterra-Neural Network (Volterra-NN), maintaining its classification performance. The obtained results show that these outputs can be used to build an array of (Volterra-NN) that needs significantly less parameters than the original array of MLPs, furthermore having the same high accuracy in most of the cases. The Volterra-NN compression capabilities have been tested by solving several kind of classification problems. Experimental results are presented on three well-known databases: Letter Recognition, Pen-Based Recognition of Handwritten Digits, and Face recognition databases. |
format |
Objeto de conferencia Objeto de conferencia |
author |
Rubiolo, Mariano |
author_facet |
Rubiolo, Mariano |
author_sort |
Rubiolo, Mariano |
title |
Extended evaluation of the Volterra-Neural Network for model compression |
title_short |
Extended evaluation of the Volterra-Neural Network for model compression |
title_full |
Extended evaluation of the Volterra-Neural Network for model compression |
title_fullStr |
Extended evaluation of the Volterra-Neural Network for model compression |
title_full_unstemmed |
Extended evaluation of the Volterra-Neural Network for model compression |
title_sort |
extended evaluation of the volterra-neural network for model compression |
publishDate |
2012 |
url |
http://sedici.unlp.edu.ar/handle/10915/123741 https://41jaiio.sadio.org.ar/sites/default/files/14_ASAI_2012.pdf |
work_keys_str_mv |
AT rubiolomariano extendedevaluationofthevolterraneuralnetworkformodelcompression |
bdutipo_str |
Repositorios |
_version_ |
1764820450161983489 |