Linear concepts and hidden variables: An empirical study

Adam J. Grove, Dan Roth

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

Abstract

Some learning techniques for classification tasks work indirectly, by first tiying to fit a full probabilistic model to the observed data. Whether this is a good idea or not depends on the robustness with respect to deviations from the postulated model. We study this question experimentally in a restricted, yet non-trivial and interesting case: we consider a conditionally independent attribute (CIA) model which postulates a single binary-valued hidden variable z on which all other attributes (i.e., the target and the observables) depend. In this model, finding the most likely value of any one variable (given known values for the others) reduces to testing a linear function of the observed values. We learn CIA with two techniques: the standard EM algorithm, and a new algorithm we develop based on covariances. We compare these, in a controlled fashion, against an algorithm (a version of Winnow) that attempts to find a good linear classifier directly. Our conclusions help delimit the fragility of using the CIA model for classification: once the data departs from this model, performance quickly degrades and drops below that of the directly-learned linear classifier.

Original languageEnglish (US)
Title of host publicationAdvances in Neural Information Processing Systems 10 - Proceedings of the 1997 Conference, NIPS 1997
PublisherNeural information processing systems foundation
Pages500-506
Number of pages7
ISBN (Print)0262100762, 9780262100762
StatePublished - Jan 1 1998
Event11th Annual Conference on Neural Information Processing Systems, NIPS 1997 - Denver, CO, United States
Duration: Dec 1 1997Dec 6 1997

Publication series

NameAdvances in Neural Information Processing Systems
ISSN (Print)1049-5258

Other

Other11th Annual Conference on Neural Information Processing Systems, NIPS 1997
CountryUnited States
CityDenver, CO
Period12/1/9712/6/97

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Information Systems
  • Signal Processing

Fingerprint Dive into the research topics of 'Linear concepts and hidden variables: An empirical study'. Together they form a unique fingerprint.

  • Cite this

    Grove, A. J., & Roth, D. (1998). Linear concepts and hidden variables: An empirical study. In Advances in Neural Information Processing Systems 10 - Proceedings of the 1997 Conference, NIPS 1997 (pp. 500-506). (Advances in Neural Information Processing Systems). Neural information processing systems foundation.