Clusterability as an Alternative to Anchor Points
When Learning with Noisy Labels
Zhaowei Zhu 1 Yiwen Song 2 Yang Liu 1
Abstract parameterized DNNs (Xia et al., 2021; Han et al., 2020),
and lead to unexpected and disparate impacts (Liu, 2021).
The label noise transition matrix, characteriz-
ing the probabilities of a training instance being A variety of approaches were proposed to a ...


雷达卡




京公网安备 11010802022788号







