Sublinear optimization for machine learning

Kenneth L. Clarkson, Elad Hazan, David P. Woodruff

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

19 Scopus citations

Abstract

We give sublinear-time approximation algorithms for some optimization problems arising in machine learning, such as training linear classifiers and finding minimum enclosing balls. Our algorithms can be extended to some kernelized versions of these problems, such as SVDD, hard margin SVM, and L 2-SVM, for which sublinear-time algorithms were not known before. These new algorithms use a combination of a novel sampling techniques and a new multiplicative update algorithm. We give lower bounds which show the running times of many of our algorithms to be nearly best possible in the unit-cost RAM model. We also give implementations of our algorithms in the semi-streaming setting, obtaining the first low pass polylogarithmic space and sublinear time algorithms achieving arbitrary approximation factor.

Original languageEnglish (US)
Title of host publicationProceedings - 2010 IEEE 51st Annual Symposium on Foundations of Computer Science, FOCS 2010
Pages449-457
Number of pages9
DOIs
StatePublished - Dec 1 2010
Externally publishedYes
Event2010 IEEE 51st Annual Symposium on Foundations of Computer Science, FOCS 2010 - Las Vegas, NV, United States
Duration: Oct 23 2010Oct 26 2010

Publication series

NameProceedings - Annual IEEE Symposium on Foundations of Computer Science, FOCS
ISSN (Print)0272-5428

Other

Other2010 IEEE 51st Annual Symposium on Foundations of Computer Science, FOCS 2010
CountryUnited States
CityLas Vegas, NV
Period10/23/1010/26/10

All Science Journal Classification (ASJC) codes

  • Computer Science(all)

Fingerprint Dive into the research topics of 'Sublinear optimization for machine learning'. Together they form a unique fingerprint.

Cite this