Sonja Greven

ICON Decomposition : Multivariate Concept-Level Explanations of Deep Representations for Model Auditing

Controlling for Omitted Variable Bias in Deep Neural Networks

This work introduces an estimation procedure that refits the final layer of a pre-trained network to include covariate effects, and shows how these effects can be orthogonalised with respect to covariates to exclude their mediated effects and that model predictions can be marginalised over the covariate distribution to...

Embedded Conditional Independence Tests for Large Language Model Generated Text with an Application to German Parliament Speeches

Embedded CITs (eCITs), which embed X and Z and apply an existing CIT to the resulting representations and to the resulting representations, are proposed and it is shown that sufficiency weakens to mean sufficiency when the embedded test targets conditional mean independence.

ICON Decomposition: Auditing Deep Neural Networks with Multivariate Variance-based Concept-level Explanations

CON decomposition is introduced, which quantifies how much of a layer's variance each concept explains given all other concepts and the outcome, and how much none of them explains, yielding layer-comparable, calibrated scores that suppress false positives.