On the generalization ability of online learning algorithms for pairwise loss functions

Purushottam Kar, Bharath Kumar Sriperumbudur, Prateek Jain, Harish C. Karnick

Research output: Contribution to conferencePaper

22 Scopus citations

Abstract

In this paper, we study the generalization properties of online learning based stochastic methods for supervised learning problems where the loss function is dependent on more than one training sample (e.g., metric learning, ranking). We present a generic decoupling technique that enables us to provide Rademacher complexity-based generalization error bounds. Our bounds are in general tighter than those obtained by Wang et al. (2012) for the same problem. Using our decoupling technique, we are further able to obtain fast convergence rates for strongly convex pairwise loss functions. We are also able to analyze a class of memory efficient online learning algorithms for pairwise learning problems that use only a bounded subset of past training samples to update the hypothesis at each step. Finally, in order to complement our generalization bounds, we propose a novel memory efficient online learning algorithm for higher order learning problems with bounded regret guarantees.

Original languageEnglish (US)
Pages1478-1486
Number of pages9
StatePublished - Jan 1 2013
Event30th International Conference on Machine Learning, ICML 2013 - Atlanta, GA, United States
Duration: Jun 16 2013Jun 21 2013

Other

Other30th International Conference on Machine Learning, ICML 2013
CountryUnited States
CityAtlanta, GA
Period6/16/136/21/13

All Science Journal Classification (ASJC) codes

  • Human-Computer Interaction
  • Sociology and Political Science

Fingerprint Dive into the research topics of 'On the generalization ability of online learning algorithms for pairwise loss functions'. Together they form a unique fingerprint.

  • Cite this

    Kar, P., Sriperumbudur, B. K., Jain, P., & Karnick, H. C. (2013). On the generalization ability of online learning algorithms for pairwise loss functions. 1478-1486. Paper presented at 30th International Conference on Machine Learning, ICML 2013, Atlanta, GA, United States.