Parallel model of online sequential extreme learning machines for classification problems with large-scale databases

Nowadays, the sizes of databases in real-world applications are around TeraByte or PetaByte. Therefore, training neural networks in reasonable times is challenging and requires high-cost computational architectures. OS-ELM is a variant of ELM, proposed for real-world applications. This algorithm al...

Descripción completa

Detalles Bibliográficos
Autores principales: Gelvez-Almeida, Elkin, Barrientos, Ricardo J., Vilches-Ponce, Karina, Mora, Marco
Formato: Objeto de conferencia
Lenguaje:Inglés
Publicado: 2023
Materias:
Ciencias Informáticas
Parallel computing
High performance computing
Extreme learning machine
Fingerprint classification
Acceso en línea:http://sedici.unlp.edu.ar/handle/10915/155423
Aporte de:
SEDICI (UNLP) de Universidad Nacional de La Plata
id I19-R120-10915-155423
record_format dspace
spelling I19-R120-10915-1554232023-07-11T20:01:39Z http://sedici.unlp.edu.ar/handle/10915/155423 isbn:978-950-34-2271-7 Parallel model of online sequential extreme learning machines for classification problems with large-scale databases Gelvez-Almeida, Elkin Barrientos, Ricardo J. Vilches-Ponce, Karina Mora, Marco 2023-06 2023 2023-07-11T17:14:20Z en Ciencias Informáticas Parallel computing High performance computing Extreme learning machine Fingerprint classification Nowadays, the sizes of databases in real-world applications are around TeraByte or PetaByte. Therefore, training neural networks in reasonable times is challenging and requires high-cost computational architectures. OS-ELM is a variant of ELM, proposed for real-world applications. This algorithm allows training with new data using the previous results without reusing the previous dataset. In this work, we present a parallel model of OS-ELM for classification problems using large-scale databases. The model consists of training several OS-ELM using multithreaded programming. The training dataset is distributed according to the number of working threads. Then, the test dataset is classified by all pre-trained OS-ELMs. Finally, the test dataset is classified using a frequency criterion. Preliminary results show that increasing the number of threads decreases the training time without significantly affecting the test accuracy of each OS-ELM. Facultad de Informática Objeto de conferencia Objeto de conferencia http://creativecommons.org/licenses/by-nc-sa/4.0/ Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) application/pdf 19-23
institution Universidad Nacional de La Plata
institution_str I-19
repository_str R-120
collection SEDICI (UNLP)
language Inglés
topic Ciencias Informáticas
Parallel computing
High performance computing
Extreme learning machine
Fingerprint classification
spellingShingle Ciencias Informáticas
Parallel computing
High performance computing
Extreme learning machine
Fingerprint classification
Gelvez-Almeida, Elkin
Barrientos, Ricardo J.
Vilches-Ponce, Karina
Mora, Marco
Parallel model of online sequential extreme learning machines for classification problems with large-scale databases
topic_facet Ciencias Informáticas
Parallel computing
High performance computing
Extreme learning machine
Fingerprint classification
description Nowadays, the sizes of databases in real-world applications are around TeraByte or PetaByte. Therefore, training neural networks in reasonable times is challenging and requires high-cost computational architectures. OS-ELM is a variant of ELM, proposed for real-world applications. This algorithm allows training with new data using the previous results without reusing the previous dataset. In this work, we present a parallel model of OS-ELM for classification problems using large-scale databases. The model consists of training several OS-ELM using multithreaded programming. The training dataset is distributed according to the number of working threads. Then, the test dataset is classified by all pre-trained OS-ELMs. Finally, the test dataset is classified using a frequency criterion. Preliminary results show that increasing the number of threads decreases the training time without significantly affecting the test accuracy of each OS-ELM.
format Objeto de conferencia
Objeto de conferencia
author Gelvez-Almeida, Elkin
Barrientos, Ricardo J.
Vilches-Ponce, Karina
Mora, Marco
author_facet Gelvez-Almeida, Elkin
Barrientos, Ricardo J.
Vilches-Ponce, Karina
Mora, Marco
author_sort Gelvez-Almeida, Elkin
title Parallel model of online sequential extreme learning machines for classification problems with large-scale databases
title_short Parallel model of online sequential extreme learning machines for classification problems with large-scale databases
title_full Parallel model of online sequential extreme learning machines for classification problems with large-scale databases
title_fullStr Parallel model of online sequential extreme learning machines for classification problems with large-scale databases
title_full_unstemmed Parallel model of online sequential extreme learning machines for classification problems with large-scale databases
title_sort parallel model of online sequential extreme learning machines for classification problems with large-scale databases
publishDate 2023
url http://sedici.unlp.edu.ar/handle/10915/155423
work_keys_str_mv AT gelvezalmeidaelkin parallelmodelofonlinesequentialextremelearningmachinesforclassificationproblemswithlargescaledatabases
AT barrientosricardoj parallelmodelofonlinesequentialextremelearningmachinesforclassificationproblemswithlargescaledatabases
AT vilchesponcekarina parallelmodelofonlinesequentialextremelearningmachinesforclassificationproblemswithlargescaledatabases
AT moramarco parallelmodelofonlinesequentialextremelearningmachinesforclassificationproblemswithlargescaledatabases
_version_ 1771439084505399296

Ejemplares similares