Classification with unknown class-conditional label noise on non-compact feature spaces
This addresses a theoretical gap in machine learning for handling label noise in non-compact spaces, with incremental contributions to understanding when finite sample rates differ from noise-free scenarios.
The paper tackles classification with unknown class-conditional label noise in non-compact feature spaces, determining minimax optimal learning rates that show a threshold behavior: rates match noise-free settings if the regression function approaches extrema quickly, but degrade otherwise, and presents an adaptive algorithm achieving these rates without prior knowledge.
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 local density. 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.