PASS: Peer-Agreement based Sample Selection for training with Noisy Labels

arXiv:2303.10802v2 Announce Type: replace
Abstract: The prevalence of noisy-label samples poses a significant challenge in deep learning, inducing overfitting effects. This has, therefore, motivated the emergence of learning with noisy-label (LNL) techniques that focus on separating noisy- and clean-label samples to apply different learning strategies to each group of samples. Current methodologies often rely on the small-loss hypothesis or feature-based selection to separate noisy- and clean-label samples, yet our empirical observations reveal their limitations, especially for labels with instance dependent …

abstract agreement apply arxiv challenge cs.cv deep learning effects emergence focus labels overfitting peer sample samples strategies training type