HIGHLIGHTS
- who: Rick Wilming from the (UNIVERSITY) have published the article: Scrutinizing XAI using linear ground-truth data with suppressor variables, in the Journal: (JOURNAL)
- what: To avoid misinterpretations due to such behavior the authors propose the actual presence of such an association as a necessary condition and objective preliminary definition for feature importance. The authors evaluate common explanation including LRP DTD PatternNet PatternAttribution LIME Anchors SHAP and permutation-based with respect to the objective definition. The authors show that most of these are unable to distinguish important features from suppressors in this setting. The authors . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.