Paper
Koh & Liang
A 2017 method proposed by Koh and Liang that introduced influence functions into deep learning, used to identify mislabeled samples and to understand model behavior.
Timeline (last 90 days)
Oct 14
Koh & Liang introduced influence functions into deep learning in 2017, using them to identify mislabeled samples and understand model behavior.
Unverified50%