The Hessian Screening Rule

Forskningsoutput: Kapitel i bok/rapport/Conference proceedingKonferenspaper i proceedingPeer review

Sammanfattning

Predictor screening rules, which discard predictors before fitting a model, have had considerable impact on the speed with which sparse regression problems, such as the lasso, can be solved. In this paper we present a new screening rule for solving the lasso path: the Hessian Screening Rule. The rule uses second-order information from the model to provide both effective screening, particularly in the case of high correlation, as well as accurate warm starts. The proposed rule outperforms all alternatives we study on simulated data sets with both low and high correlation for `1-regularized least-squares (the lasso) and logistic regression. It also performs best in general on the real data sets that we examine.

Originalspråkengelska
Titel på värdpublikationAdvances in Neural Information Processing Systems
RedaktörerS. Koyejo, S. Mohamed, A. Agarwal, D. Belgrave, K. Cho, A. Oh
FörlagCurran Associates, Inc
Sidor25404-25421
Volym35
ISBN (elektroniskt)9781713871088
StatusPublished - 2022 dec. 6
Evenemang36th Conference on Neural Information Processing Systems, NeurIPS 2022 - New Orleans, USA
Varaktighet: 2022 nov. 282022 dec. 9

Publikationsserier

NamnAdvances in Neural Information Processing Systems
Volym35
ISSN (tryckt)1049-5258

Konferens

Konferens36th Conference on Neural Information Processing Systems, NeurIPS 2022
Land/TerritoriumUSA
OrtNew Orleans
Period2022/11/282022/12/09

Bibliografisk information

Publisher Copyright:
© 2022 Neural information processing systems foundation. All rights reserved.

Ämnesklassifikation (UKÄ)

  • Sannolikhetsteori och statistik

Fingeravtryck

Utforska forskningsämnen för ”The Hessian Screening Rule”. Tillsammans bildar de ett unikt fingeravtryck.

Citera det här