156 citations · 179 across the 2 of their papers we have counts for
3 papers · 1 filter
Manifold Restricted Interventional Shapley Values
Muhammad Faaiz Taufiq, Patrick Blöbaum, Lenon Minorics
Shapley values are model-agnostic methods for explaining model predictions. Many commonly used methods of computing Shapley values, known as off-manifold methods, rely on model eva…
Causal structure based root cause analysis of outliers
Dominik Janzing, Kailash Budhathoki, Lenon Minorics +1
We describe a formal approach to identify 'root causes' of outliers observed in variables in a scenario where the causal relation between the variables is a kno…
Feature relevance quantification in explainable AI: A causal problem
Dominik Janzing, Lenon Minorics, Patrick Blöbaum
We discuss promising recent contributions on quantifying feature relevance using Shapley values, where we observed some confusion on which probability distribution is the right one…