no code implementations • ICLR 2022 • Julius Adebayo, Michael Muelly, Hal Abelson, Been Kim
We investigate whether three types of post hoc model explanations--feature attribution, concept activation, and training point ranking--are effective for detecting a model's reliance on spurious signals in the training data.
1 code implementation • NeurIPS 2020 • Julius Adebayo, Michael Muelly, Ilaria Liccardi, Been Kim
For several explanation methods, we assess their ability to: detect spurious correlation artifacts (data contamination), diagnose mislabeled training examples (data contamination), differentiate between a (partially) re-initialized model and a trained one (model contamination), and detect out-of-distribution inputs (test-time contamination).
1 code implementation • ICCV 2019 • Lanlan Liu, Michael Muelly, Jia Deng, Tomas Pfister, Li-Jia Li
This paper explores object detection in the small data regime, where only a limited number of annotated bounding boxes are available due to data rarity and annotation expense.
5 code implementations • NeurIPS 2018 • Julius Adebayo, Justin Gilmer, Michael Muelly, Ian Goodfellow, Moritz Hardt, Been Kim
We find that reliance, solely, on visual assessment can be misleading.