Jun 10, 2022 · Abstract:Counterfactuals can explain classification decisions of neural networks in a human interpretable way.

Apr 3, 2024 · We analyze the generation process theoretically using Riemannian differential geometry and validate the quality of the generated counterfactuals.

We propose two methods to leverage generative models to construct such suitable coordinate systems that are either exactly or approximately diffeomorphic.

This work performs a suitable diffeomorphic coordinate transformation and then performs gradient ascent in these coordinates to find counterfactuals which ...

We propose two methods to leverage generative models to construct such suitable coordinate systems that are either exactly or approximately diffeomorphic. We ...

Jun 10, 2022 · We analyze the generation process theoretically using Riemannian differential geometry and validate the quality of the generated counterfactuals ...

We describe CounterSynth, a conditional generative model of diffeomorphic deformations that induce label-driven, biologically plausible changes in volumetric ...

Diffeomorphic Counterfactuals with Generative Models ... Counterfactuals can explain classification decisions of neural networks in a human interpretable way. 12.

We create counterfactual explanations for image data by doing gradient ascent in the latent space of a generative model. Our method finds counterfactuals ...

Nov 4, 2024 · Diffeomorphic counterfactuals with generative models. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023. Domnich and ...