Sparse Counterfactual Explanation

Sparse counterfactual explanations aim to understand model predictions by identifying minimal changes to input data that alter the model's output. Current research focuses on developing algorithms, such as genetic algorithms and generative adversarial networks, to generate these sparse explanations across various data types, including time series and images, while addressing challenges like ensuring validity, plausibility, and robustness to noise. This work is crucial for improving the transparency and trustworthiness of machine learning models, particularly in high-stakes applications where understanding model decisions is paramount. The development of robust and easily interpretable sparse counterfactual explanations is driving progress towards more explainable AI.

Papers

September 14, 2024

TX-Gen: Multi-Objective Optimization for Sparse Counterfactual Explanations for Time-Series Classification
Qi Huang, Sofoklis Kitharidis, Thomas Bäck, Niki van Stein
Time Series Counterfactual Explanation Multi Objective Optimization Time Series Classification High Quality Counterfactuals Sparse Counterfactual Explanation

September 29, 2022

Local and Regional Counterfactual Rules: Summarized and Robust Recourses
Salim I. Amoukou, Nicolas J. B Brunel
Counterfactual Explanation Robust Recourse Counterfactual Plan Sparse Counterfactual Explanation

June 2, 2022

Generating Sparse Counterfactual Explanations For Multivariate Time Series
Jana Lang, Martin Giese, Winfried Ilg, Sebastian Otte
Generative Adversarial Network Counterfactual Explanation Multivariate Time Series Counterfactual Prediction Synthetic Time Series Sparse Counterfactual Explanation

May 16, 2022

Sparse Visual Counterfactual Explanations in Image Space
Valentyn Boreiko, Maximilian Augustin, Francesco Croce, Philipp Berens, Matthias Hein
Adversarial Example ImageNet Dataset Image Classifier Visual Counterfactual Explanation ImageNet Classifier Image Space Sparse Counterfactual Explanation

January 22, 2022

On the Robustness of Sparse Counterfactual Explanations to Adverse Perturbations
Marco Virgolin, Saverio Fracaros
Native Robustness Counterfactual Explanation Explainable Machine Learning Unintentional Perturbation Sparse Counterfactual Explanation

Sparse Counterfactual Explanation

Papers

TX-Gen: Multi-Objective Optimization for Sparse Counterfactual Explanations for Time-Series Classification

Local and Regional Counterfactual Rules: Summarized and Robust Recourses

Generating Sparse Counterfactual Explanations For Multivariate Time Series

Sparse Visual Counterfactual Explanations in Image Space

On the Robustness of Sparse Counterfactual Explanations to Adverse Perturbations