L2B: Learning to Bootstrap Robust Models for Combating Label Noise

arXiv:2202.04291v2 Announce Type: replace
Abstract: Deep neural networks have shown great success in representation learning. However, when learning with noisy labels (LNL), they can easily overfit and fail to generalize to new data. This paper introduces a simple and effective method, named Learning to Bootstrap (L2B), which enables models to bootstrap themselves using their own predictions without being adversely affected by erroneous pseudo-labels. It achieves this by dynamically adjusting the importance weight between real observed and generated labels, as well …

abstract arxiv bootstrap cs.cv data however labels networks neural networks noise paper representation representation learning robust robust models simple success type