Distributed mixture-of-experts for Big Data using PETUUM framework

Billy Peralta, Luis Parra, Oriel Herrera, Luis Caro

Resultado de la investigación: Contribución a los tipos de informe/libroContribución a la conferenciarevisión exhaustiva

Resumen

Today, organizations are beginning to realize the importance of using as much data as possible for decision-making in their strategy. The finding of relevant patterns in enormous amount of data requires automatic machine learning algorithms, among them, a popular option is the mixture-of-experts that allows to model data using a set of local experts. The problem of using typical learning algorithms over Big Data is the handling of these large datasets in primary memory. In this paper, we propose a methodology to learn a mixture-of-experts in a distributed way using PETUUM platform. Particularly, we propose to learn the parameters of mixture-of-experts by adapting the standard stochastic gradient descent in a distributed way. This methodology is applied to people detection with standard real datasets considering accuracy and precision metrics among other. The results show a consistent performance of mixture-of-experts models where the best number of experts varies according to the particular dataset. We also evidence the advantages of the distributed approach by showing the almost linear decreasing of average training time according to the number of processors. In a future work, we expect to apply this methodology to mixture-of-experts with embedded variable selection.

Idioma originalInglés
Título de la publicación alojada2017 36th International Conference of the Chilean Computer Science Society, SCCC 2017
EditorialIEEE Computer Society
Páginas1-7
Número de páginas7
Volumen2017-October
ISBN (versión digital)9781538634837
DOI
EstadoPublicada - 5 jul 2018
Evento36th International Conference of the Chilean Computer Science Society, SCCC 2017 - Arica, Chile
Duración: 16 oct 201720 oct 2017

Serie de la publicación

NombreProceedings - International Conference of the Chilean Computer Science Society, SCCC
Volumen2017-October
ISSN (versión impresa)1522-4902

Conferencia

Conferencia36th International Conference of the Chilean Computer Science Society, SCCC 2017
País/TerritorioChile
CiudadArica
Período16/10/1720/10/17

Áreas temáticas de ASJC Scopus

  • Ingeniería (todo)
  • Informática (todo)

Huella

Profundice en los temas de investigación de 'Distributed mixture-of-experts for Big Data using PETUUM framework'. En conjunto forman una huella única.

Citar esto