Learning noisy linear classifiers via adaptive and selective sampling

doi:10.1007/s10994-010-5191-x

Learning noisy linear classifiers via adaptive and selective sampling

2010-05-20

21

OA

AI

G

Giovanni Cavallanti *

DOI：10.1007/s10994-010-5191-x

Original

Original request for help

Save

Abstract

En 中文

We introduce efficient margin-based algorithms for selective sampling and filtering in binary classification tasks. Experiments on real-world textual data reveal that our algorithms perform significantly better than popular and similarly efficient competitors. Using the so-called Mammen-Tsybakov low noise condition to parametrize the instance distribution, and assuming linear label noise, we show bounds on the convergence rate to the Bayes risk of a weaker adaptive variant of our selective sampler. Our analysis reveals that, excluding logarithmic factors, the average risk of this adaptive sampler converges to the Bayes risk at rate N (-(1+alpha)(2+alpha)/2(3+alpha)) where N denotes the number of queried labels, and alpha > 0 is the exponent in the low noise condition. For all this convergence rate is asymptotically faster than the rate N (-(1+alpha)/(2+alpha)) achieved by the fully supervised version of the base selective sampler, which queries all labels. Moreover, for alpha -> a (hard margin condition) the gap between the semi- and fully-supervised rates becomes exponential.

Keywords:

Active learning

Selective sampling

Adaptive sampling

Linear classification

Low noise

Learning noisy linear classifiers via adaptive and selective sampling

Abstract

Journal

Organization