Diffeomorphic Counterfactuals with Generative Models

TL;DR

This paper introduces a method for generating human-interpretable counterfactual explanations for neural network decisions by applying diffeomorphic transformations and gradient ascent in transformed coordinates, leveraging generative models.

Contribution

It presents a novel approach combining diffeomorphic coordinate transformations with generative models to produce high-quality counterfactuals for neural network interpretability.

Findings

Counterfactuals are classified with high confidence as target classes.

The method effectively leverages diffeomorphic transformations for counterfactual generation.

Qualitative and quantitative analyses validate the approach's effectiveness.

Abstract

Counterfactuals can explain classification decisions of neural networks in a human interpretable way. We propose a simple but effective method to generate such counterfactuals. More specifically, we perform a suitable diffeomorphic coordinate transformation and then perform gradient ascent in these coordinates to find counterfactuals which are classified with great confidence as a specified target class. We propose two methods to leverage generative models to construct such suitable coordinate systems that are either exactly or approximately diffeomorphic. We analyze the generation process theoretically using Riemannian differential geometry and validate the quality of the generated counterfactuals using various qualitative and quantitative measures.