Fast training for Large-Scale One-versus-All linear classifiers using Tree-Structured initialization

Huang Fang, Minhao Cheng, Cho Jui Hsieh, Michael Friedlander

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Scopus citations

Abstract

We consider the problem of training one-versus-all (OVA) linear classifiers for multiclass or multilabel classification when the number of labels is large. A naive extension of OVA to this problem, even with hundreds of cores, usually requires hours for training on large real world datasets. We propose a novel algorithm called OVA-Primal++ that speeds up the training of OVA by using a tree-structured training order, where each classifier is trained using its parent’s classifier as initialization. OVA-Primal++ is both theoretically and empirically faster than the naive OVA algorithm, and yet still enjoys the same highly parallelizability and small memory footprint. Extensive experiments on multiclass and multilabel classification datasets validate the effectiveness of our method.

Original languageEnglish (US)
Title of host publicationSIAM International Conference on Data Mining, SDM 2019
PublisherSociety for Industrial and Applied Mathematics Publications
Pages280-288
Number of pages9
ISBN (Electronic)9781611975673
DOIs
StatePublished - 2019
Event19th SIAM International Conference on Data Mining, SDM 2019 - Calgary, Canada
Duration: May 2 2019May 4 2019

Publication series

NameSIAM International Conference on Data Mining, SDM 2019

Conference

Conference19th SIAM International Conference on Data Mining, SDM 2019
Country/TerritoryCanada
CityCalgary
Period5/2/195/4/19

All Science Journal Classification (ASJC) codes

  • Software

Cite this