Positive-Congruent Training: Towards Regression-Free Model Updates

Sijie Yan
Sijie Yan
Kaustav Kundu
Kaustav Kundu
Shuo Yang
Shuo Yang
Siqi Deng
Siqi Deng
Meng Wang
Meng Wang
Wei Xia
Wei Xia
Cited by: 0|Views11

Abstract:

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) ...More

Code:

Data:

Full Text
Bibtex
Your rating :
0

 

Tags
Comments