REAL@UCSC

Human annotated noisy labels for CIFAR-10 and CIFAR-100. The website of CIFAR-N is available at http://www.noisylabels.com/.

UCSC-REAL

A Second-Order Approach to Learning with Instance-Dependent Label Noise (CVPR'21 oral)

UCSC-REAL

Learning with Instance-Dependent Label Noise: A Sample Sieve Approach (ICLR2021)

UCSC-REAL

PyTorch Implementation of ECCV 2022 paper: "DuelGAN: A Duel Between Two Discriminators Stabilizes the GAN Training".

UCSC-REAL

UCSC-REAL

[ICML2022 Long Talk] Official Pytorch implementation of "To Smooth or Not? When Label Smoothing Meets Noisy Labels"

UCSC-REAL

UCSC-REAL