Classification with unknown class-conditional label noise on non-compact feature spaces

Research output: Contribution to journalArticle (Academic Journal)

1 Downloads (Pure)

Abstract

We investigate the problem of classification in the presence of unknown class-conditional label noise in which the labels observed by the learner have been corrupted with some unknown class dependent probability. In order to obtain finite sample rates, previous approaches to classification with unknown class-conditional label noise have required that the regression function is close to its extrema on sets of large measure. We shall consider this problem in the setting of non-compact metric spaces, where the regression function need not attain its extrema. In this setting we determine the minimax optimal learning rates (up to logarithmic factors). The rate displays interesting threshold behaviour: When the regression function approaches its extrema at a sufficient rate, the optimal learning rates are of the same order as those obtained in the label-noise free setting. If the regression function approaches its extrema more gradually then classification performance necessarily degrades. In addition, we present an adaptive algorithm which attains these rates without prior knowledge of either the distributional parameters or the localdensity. This identifies for the first time a scenario in which finite sample rates are achievable in the label noise setting, but they differ from the optimal rates without label noise..
Original languageEnglish
Pages (from-to)2624-2651
Number of pages28
JournalProceedings of Machine Learning Research
Volume99
Publication statusPublished - 17 Aug 2019
EventConference on Learning Theory - Phoenix, United States
Duration: 25 Jun 201928 Jun 2019
Conference number: 32
https://learningtheory.org/colt2019/

Keywords

  • Label noise
  • minimax rates
  • non-parametric classification
  • metric spaces

Fingerprint Dive into the research topics of 'Classification with unknown class-conditional label noise on non-compact feature spaces'. Together they form a unique fingerprint.

Cite this