Learning With Noisy Labels On Cifar 10N

Metrics

Accuracy (mean)

Results

Performance results of various models on this benchmark

		Paper Title
ProMix	97.39	ProMix: Combating Label Noise via Maximizing Clean Sample Utility
PSSCL	96.41	PSSCL: A progressive sample selection framework with contrastive loss designed for noisy labels
PGDF	96.11	Sample Prior Guided Robust Model Learning to Suppress Noisy Labels
SOP+	95.61	Robust Training under Label Noise by Over-parameterization
ILL	95.47	Imprecise Label Learning: A Unified Framework for Learning with Various Imprecise Label Configurations
CORES*	95.25	Learning with Instance-Dependent Label Noise: A Sample Sieve Approach
Divide-Mix	95.01	DivideMix: Learning with Noisy Labels as Semi-supervised Learning
ELR+	94.83	Early-Learning Regularization Prevents Memorization of Noisy Labels
PES (Semi)	94.66	Understanding and Improving Early Stopping for Learning with Noisy Labels
GNL	92.57	Partial Label Supervision for Agnostic Generative Noisy Label Learning
ELR	92.38	Early-Learning Regularization Prevents Memorization of Noisy Labels
CAL	91.97	Clusterability as an Alternative to Anchor Points When Learning with Noisy Labels
Negative-LS	91.97	To Smooth or Not? When Label Smoothing Meets Noisy Labels
F-div	91.64	When Optimizing $f$-divergence is Robust with Label Noise
Positive-LS	91.57	Does label smoothing mitigate label noise?
JoCoR	91.44	Combating noisy labels by agreement: A joint training method with co-regularization
CORES	91.23	Learning with Instance-Dependent Label Noise: A Sample Sieve Approach
Co-Teaching	91.20	Co-teaching: Robust Training of Deep Neural Networks with Extremely Noisy Labels
Peer Loss	90.75	Peer Loss Functions: Learning from Noisy Labels without Knowing Noise Rates
Co-Teaching+	90.61	How does Disagreement Help Generalization against Label Corruption?

0 of 26 row(s) selected.

Command Palette

Learning With Noisy Labels On Cifar 10N

Metrics

Results