Label-noise robust logistic regression (rLR) is an extension of logistic regression that includes a model of random mislabelling. This paper attempts a theoretical analysis of rLR. By decomposing and interpreting the gradient of the likelihood objective of rLR as employed in gradient ascent optimisation, we get insights into the ability of the rLR learning algorithm to counteract the negative effect of mislabelling as a result of an intrinsic re-weighting mechanism. We also give an upper-bound on the error of rLR using Rademacher complexities. © 2013 Springer-Verlag.
CITATION STYLE
Bootkrajang, J., & Kabán, A. (2013). Learning a label-noise robust logistic regression: Analysis and experiments. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8206 LNCS, pp. 569–576). https://doi.org/10.1007/978-3-642-41278-3_69
Mendeley helps you to discover research relevant for your work.