Por favor, use este identificador para citar o enlazar este ítem: https://hdl.handle.net/20.500.12008/51581

Título:

Multi-task supervised learning via cross-learning

Otros títulos:

Coordinador:

Director:

Compilador:

Autor:

Cerviño, Juan
Bazerque, Juan Andrés
Calvo-Fullana, Miguel
Ribeiro, Alejandro

Tutor:

Tipo de documento:

Preprint

Editor:

Palabras clave:

Supervised learning
Multi-task learning
Optimization
Fitting
Neural networks
Signal processing algorithms
Europe
Signal processing
Gaussian distribution

Descriptores:

Año de publicación:

2021

Contenido:

Resumen:

In this paper we consider a problem known as multi-task learning, consisting of fitting a set of classifier or regression functions intended for solving different tasks. In our novel formulation, we couple the parameters of these functions, so that they learn in their task specific domains while staying close to each other. This facilitates cross-fertilization in which data collected across different domains help improving the learning performance at each other task. First, we present a simplified case in which the goal is to estimate the means of two Gaussian variables, for the purpose of gaining some insights on the advantage of the proposed cross-learning strategy. Then we provide a stochastic projected gradient algorithm to perform cross-learning over a generic loss function. If the number of parameters is large, then the projection step becomes computationally expensive. To avoid this situation, we derive a primal-dual algorithm that exploits the structure of the dual problem, achieving a formulation whose complexity only depends on the number of tasks. Preliminary numerical experiments for image classification by neural networks trained on a dataset divided in different domains corroborate that the cross-learned function outperforms both the task-specific and the consensus approaches

Descripción:

metadata.articulos.dc.description.uri:

Editorial:

EN:

Financiadores:

NSF-Simons MoDLTheorinet
ANII FSE 1-2019-1-157459

Citación:

Cerviño, J., Bazerque, J., Calvo-Fullana, M. y otros. Multi-task supervised learning via cross-learning [Preprint]. Publicado en: 2021 29th European Signal Processing Conference (EUSIPCO), Dublin, Ireland, 23-27 aug. 2021, pp. 1381-1385.

Citación:

metadata.articulos.cc.license.name:

ISBN:

e-ISBN:

ISSN:

ISMN:

Otros identificadores:

Cobertura geográfica:

Cobertura temporal:

Departamento académico:

Sistemas y Control

Grupo de investigación:

Licencia:

Licencia Creative Commons Atribución (CC - By 4.0)

Colecciones:

Publicaciones académicas y científicas
FicherosDescripciónTamañoFormato
CBCR21a.pdf Preprint 2.37 MB Adobe PDF
Mostrar el registro Dublin Core completo del ítem