no code implementations • 1 Jan 2021 • Sheng-Min Shih, Pin-Ju Tien, Zohar Karnin
Our approach effectively selects the baseline as the closest realistic sample belong to the target class, which allows attribution methods to provide true one-vs-one explanations.
1 code implementation • 11 Nov 2020 • Sheng-Min Shih, Pin-Ju Tien, Zohar Karnin
Attribution methods have been shown as promising approaches for identifying key features that led to learned model predictions.