Improving calibration for long-tailed recognition

Z Zhong, J Cui, S Liu, J Jia - … of the IEEE/CVF conference on …, 2021 - openaccess.thecvf.com
Deep neural networks may perform poorly when training datasets are heavily class-
imbalanced. Recently, two-stage methods decouple representation learning and classifier …
被引用次数：299 相关文章
[PDF] 47.243.74.31

[PDF][PDF] Improving Calibration for Long-Tailed Recognition

Z Zhong, J Cui, S Liu, J Jia - 47.243.74.31
Deep neural networks may perform poorly when training datasets are heavily class-
imbalanced. Recently, twostage methods decouple representation learning and classifier …

Improving Calibration for Long-Tailed Recognition

Z Zhong, J Cui, S Liu, J Jia - openreview.net
Deep neural networks often perform poorly when training datasets are heavily class-
imbalanced. Recently, two-stage methods greatly improve the performances by decoupling …

Improving Calibration for Long-Tailed Recognition

Z Zhong, J Cui, S Liu, J Jia - arXiv e-prints, 2021 - ui.adsabs.harvard.edu
Deep neural networks may perform poorly when training datasets are heavily class-
imbalanced. Recently, two-stage methods decouple representation learning and classifier …

Improving Calibration for Long-Tailed Recognition

Z Zhong, J Cui, S Liu, J Jia - arXiv preprint arXiv:2104.00466, 2021 - arxiv.org
Deep neural networks may perform poorly when training datasets are heavily class-
imbalanced. Recently, two-stage methods decouple representation learning and classifier …

Improving Calibration for Long-Tailed Recognition

Z Zhong, J Cui, S Liu, J Jia - 2021 IEEE/CVF Conference on …, 2021 - computer.org
Deep neural networks may perform poorly when training datasets are heavily class-
imbalanced. Recently, two-stage methods decouple representation learning and classifier …

Improving Calibration for Long-Tailed Recognition

Z Zhong, J Cui, S Liu, J Jia - 2021 IEEE/CVF Conference on …, 2021 - ieeexplore.ieee.org
Deep neural networks may perform poorly when training datasets are heavily class-
imbalanced. Recently, two-stage methods decouple representation learning and classifier …
[PDF] jiaya.me

[PDF][PDF] Improving Calibration for Long-Tailed Recognition

Z Zhong, J Cui, S Liu, J Jia - jiaya.me
Deep neural networks may perform poorly when training datasets are heavily class-
imbalanced. Recently, twostage methods decouple representation learning and classifier …