Part-dependent Label Noise: Towards Instance-dependent Label Noise

By Xiaobo Xia, Tongliang Liu, Bo Han, Nannan Wang, Mingming Gong, Haifeng Liu, Gang Niu, Dacheng Tao, Masashi Sugiyama. Neural Information Processing Systems, 2020

313 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/e9216c9b4dbd93551de772ddc8f4d9fd123e9f85

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Xiaobo Xia
2. Bo Han
3. Tongliang Liu
4. Ying-Long Bai
5. Gang Niu
6. Chen Gong
7. Nannan Wang
8. Erkun Yang

Citations in the corpus

• Understanding and Improving Early Stopping for Learning with Noisy Labels (2021, background): “Methods in the first category mainly model noisy labels with the noise transition matrix [26, 37, 36, 32].”
• Beyond Class-Conditional Assumption: A Primary Attempt to Combat Instance-Dependent Label Noise (2020, unknown): “Some pioneer efforts has been contributed to IDN, but most results are restricted to binary classification (Menon, van Rooyen, and Natarajan 2018; Bootkrajang and Chaijaruwanich 2018; Cheng et al. 2020) or based on assumptions such as the noise is parts-dependent (Xia et al. 2020).”
• Robust early-learning: Hindering the memorization of noisy labels (2021, methodology): “Following (Xia et al., 2020b), we generate this type of label noise to validate the effectiveness of the proposed method.”