Tag: label noise

Researchers show that conformal prediction set sizes and miscoverage changes can distinguish label noise from covariate shift in deployed classifiers ...

Researchers have developed a transition matrix-based framework that identifies the target classes of backdoor attacks in neural networks and sharply ...

Researchers have developed SLAW, a training framework that adapts label smoothing to the sharpness of the loss landscape and reweights ...

Researchers have developed CLMSynth, an open-source Python package that generates synthetic labels with precisely controlled agreement to known cluster structures, ...

A new audit protocol reveals that machine learning models trained on cheap proxy labels for cryptocurrency risk can excel at ...

Researchers have introduced PMGL-GAD, a framework that combines training-free green learning features with a global attention disambiguation module to solve ...