Learning with Imbalanced Noisy Data by Preventing Bias in Sample Selection

arXiv:2402.11242v1 Announce Type: new
Abstract: Learning with noisy labels has gained increasing attention because the inevitable imperfect labels in real-world scenarios can substantially hurt the deep model performance. Recent studies tend to regard low-loss samples as clean ones and discard high-loss ones to alleviate the negative impact of noisy labels. However, real-world datasets contain not only noisy labels but also class imbalance. The imbalance issue is prone to causing failure in the loss-based sample selection since the under-learning of tail …

abstract arxiv attention bias cs.ai cs.lg data impact labels loss low negative performance regard sample samples studies type world