Attribute noise robust binary classification
This work addresses robustness in binary classification for noisy features, but it is incremental as it focuses on specific noise models and loss functions.
The paper tackles the problem of learning linear classifiers with binary features and labels under attribute noise, showing that squared loss is robust in the Sy-De noise model but not in the Asy-In model, with empirical support for low to moderate noise rates.
We consider the problem of learning linear classifiers when both features and labels are binary. In addition, the features are noisy, i.e., they could be flipped with an unknown probability. In Sy-De attribute noise model, where all features could be noisy together with same probability, we show that $0$-$1$ loss ($l_{0-1}$) need not be robust but a popular surrogate, squared loss ($l_{sq}$) is. In Asy-In attribute noise model, we prove that $l_{0-1}$ is robust for any distribution over 2 dimensional feature space. However, due to computational intractability of $l_{0-1}$, we resort to $l_{sq}$ and observe that it need not be Asy-In noise robust. Our empirical results support Sy-De robustness of squared loss for low to moderate noise rates.