Learning with many experts: model selection and sparsity
This work addresses model selection challenges in noisy label settings, which is an incremental improvement for machine learning practitioners dealing with imprecise expert annotations.
The paper tackles the problem of model selection for classifiers trained with noisy expert labels when true labels are unavailable, by introducing a surrogate loss with theoretical consistency guarantees and applying it to tune sparsity-inducing penalization, demonstrating effectiveness on simulated and real datasets.
Experts classifying data are often imprecise. Recently, several models have been proposed to train classifiers using the noisy labels generated by these experts. How to choose between these models? In such situations, the true labels are unavailable. Thus, one cannot perform model selection using the standard versions of methods such as empirical risk minimization and cross validation. In order to allow model selection, we present a surrogate loss and provide theoretical guarantees that assure its consistency. Next, we discuss how this loss can be used to tune a penalization which introduces sparsity in the parameters of a traditional class of models. Sparsity provides more parsimonious models and can avoid overfitting. Nevertheless, it has seldom been discussed in the context of noisy labels due to the difficulty in model selection and, therefore, in choosing tuning parameters. We apply these techniques to several sets of simulated and real data.