Speeding-up one-versus-all training for extreme classification via mean-separating initialization
Loading...
Access rights
openAccess
publishedVersion
URL
Journal Title
Journal ISSN
Volume Title
A4 Artikkeli konferenssijulkaisussa
This publication is imported from Aalto University research portal.
View publication in the Research portal (opens in new window)
View/Open full text file from the Research portal (opens in new window)
Other link related to publication (opens in new window)
View publication in the Research portal (opens in new window)
View/Open full text file from the Research portal (opens in new window)
Other link related to publication (opens in new window)
Authors
Date
2022-11
Department
Major/Subject
Mcode
Degree programme
Language
en
Pages
Series
Machine Learning, Volume 111, issue 11, pp. 3953-3976
Abstract
In this paper, we show that a simple, data dependent way of setting the initial vector can be used to substantially speed up the training of linear one-versus-all classifiers in extreme multi-label classification (XMC). We discuss the problem of choosing the initial weights from the perspective of three goals. We want to start in a region of weight space (a) with low loss value, (b) that is favourable for second-order optimization, and (c) where the conjugate-gradient (CG) calculations can be performed quickly. For margin losses, such an initialization is achieved by selecting the initial vector such that it separates the mean of all positive (relevant for a label) instances from the mean of all negatives – two quantities that can be calculated quickly for the highly imbalanced binary problems occurring in XMC. We demonstrate a training speedup of up to 5× on Amazon-670K dataset with 670,000 labels. This comes in part from the reduced number of iterations that need to be performed due to starting closer to the solution, and in part from an implicit negative-mining effect that allows to ignore easy negatives in the CG step. Because of the convex nature of the optimization problem, the speedup is achieved without any degradation in classification accuracy. The implementation can be found at https://github.com/xmc-aalto/dismecpp.Description
Keywords
Large-scale multi-label classification, Linear classification, 2nd order optimization, Class imbalance, Weight initialization
Other note
Citation
Schultheis, E & Babbar, R 2022, ' Speeding-up one-versus-all training for extreme classification via mean-separating initialization ', Machine Learning, vol. 111, no. 11, pp. 3953-3976 . https://doi.org/10.1007/s10994-022-06228-2