Do Users Benefit From Interpretable Vision? A User Study, Baseline, And Dataset

1 code implementation • ICLR 2022 • Leon Sixt, Martin Schuessler, Oana-Iuliana Popescu, Philipp Weiß, Tim Landgraf

In a study, we assess if participants can identify the relevant set of attributes compared to the ground-truth.

counterfactual Image Classification

Two4Two: Evaluating Interpretable Machine Learning - A Synthetic Dataset For Controlled Experiments

1 code implementation • 6 May 2021 • Martin Schuessler, Philipp Weiß, Leon Sixt

However, few of these approaches are subjected to human-subject evaluations, partly because it is challenging to design controlled experiments with natural image datasets, as they leave essential factors out of the researcher's control.

BIG-bench Machine Learning Image Classification +1

Interpretability Through Invertibility: A Deep Convolutional Network With Ideal Counterfactuals And Isosurfaces

no code implementations • 1 Jan 2021 • Leon Sixt, Martin Schuessler, Philipp Weiß, Tim Landgraf

Using PCA on the classifier’s input, we can also create “isofactuals”– image interpolations with the same outcome but visually meaningful different features.