Positive-Congruent Training: Towards Regression-Free Model Updates
Positive-Congruent Training: Towards Regression-Free Model Updates
复制标题
正一致性训练：迈向无回归模型更新
DOI：
10.1109/cvpr46437.2021.01407
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Stefano Soatto
中科院分区：
文献类型：
--
作者：
Sijie Yan;Yuanjun Xiong;Kaustav Kundu;Shuo Yang;Siqi Deng;Meng Wang;Wei Xia;Stefano Soatto
减少AI系统不同版本的行为的不一致与减少图像分类的整体误差一样重要。通过旧（参考）模型正确分类。与模型蒸馏不同，我们与参考模型的一致性。参考模型本身可以作为多个深神经网络的集合，可以进一步降低负面额，而不会影响新模型的准确性。
Reducing inconsistencies in the behavior of different versions of an AI system can be as important in practice as reducing its overall error. In image classification, sample-wise inconsistencies appear as "negative flips": A new model incorrectly predicts the output for a test sample that was correctly classified by the old (reference) model. Positive-congruent (PC) training aims at reducing error rate while at the same time reducing negative flips, thus maximizing congruency with the reference model only on positive predictions, unlike model distillation. We propose a simple approach for PC training, Focal Distillation, which enforces congruence with the reference model by giving more weights to samples that were correctly classified. We also found that, if the reference model itself can be chosen as an ensemble of multiple deep neural networks, negative flips can be further reduced without affecting the new model’s accuracy.
影响因子：
5.4
作者：
RATCLIFF, R
通讯作者：
RATCLIFF, R