Sciweavers

ICML
2001
IEEE

Estimating a Kernel Fisher Discriminant in the Presence of Label Noise

15 years 7 days ago
Estimating a Kernel Fisher Discriminant in the Presence of Label Noise
Data noise is present in many machine learning problems domains, some of these are well studied but others have received less attention. In this paper we propose an algorithm for constructing a kernel Fisher discriminant (KFD) from training examples with noisy labels. The approach allows to associate with each example a probability of the label being flipped. We utilise an expectation maximization (EM) algorithm for updating the probabilities. The E-step uses class conditional probabilities estimated as a by-product of the KFD algorithm. The M-step updates the flip probabilities and determines the parameters of the discriminant. We demonstrate the feasibility of the approach on two real-world data-sets.
Bernhard Schölkopf, Neil D. Lawrence
Added 17 Nov 2009
Updated 17 Nov 2009
Type Conference
Year 2001
Where ICML
Authors Bernhard Schölkopf, Neil D. Lawrence
Comments (0)