Peer Loss Functions: Learning from Noisy Labels without Knowing Noise Rates
This simplifies model development for practitioners dealing with noisy training labels by eliminating the need to specify or estimate noise rates.
The paper tackles learning from noisy labels without requiring prior knowledge of noise rates by introducing peer loss functions, showing that empirical risk minimization with these functions yields optimal or near-optimal classifiers comparable to using clean data, with experimental validation.
Learning with noisy labels is a common challenge in supervised learning. Existing approaches often require practitioners to specify noise rates, i.e., a set of parameters controlling the severity of label noises in the problem, and the specifications are either assumed to be given or estimated using additional steps. In this work, we introduce a new family of loss functions that we name as peer loss functions, which enables learning from noisy labels and does not require a priori specification of the noise rates. Peer loss functions work within the standard empirical risk minimization (ERM) framework. We show that, under mild conditions, performing ERM with peer loss functions on the noisy dataset leads to the optimal or a near-optimal classifier as if performing ERM over the clean training data, which we do not have access to. We pair our results with an extensive set of experiments. Peer loss provides a way to simplify model development when facing potentially noisy training labels, and can be promoted as a robust candidate loss function in such situations.