CounteRGAN: Generating counterfactuals for real-time recourse and interpretability using residual GANs

Daniel Nemirovsky, Nicolas Thiebaut, Ye Xu, Abhishek Gupta
Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence, PMLR 180:1488-1497, 2022.

Abstract

Model interpretability, fairness, and recourse for end users have increased as machine learning models have become increasingly popular in areas including criminal justice, finance, healthcare, and job marketplaces. This work presents a novel method of addressing these issues by producing meaningful counterfactuals that are aimed at providing recourse to users and highlighting potential model biases. A meaningful counterfactual is a reasonable alternative scenario that illustrates how input data perturbations can influence the model’s output. The CounteRGAN method generates meaningful counterfactuals for a target classifier by utilizing a novel Residual Generative Adversarial Network (RGAN). We compare our method against leading state-of-the-art approaches on image and tabular datasets over a variety of performance metrics. The results indicate a significant improvement over existing techniques in combined metric performance, with a latency reduction of 2 to 7 orders of magnitude which enables providing real-time recourse to users. The code for reproducibility can be found here: https://github.com/gan-counterfactuals/countergan.

Cite this Paper

BibTeX
@InProceedings{pmlr-v180-nemirovsky22a, title = {CounteRGAN: Generating counterfactuals for real-time recourse and interpretability using residual GANs}, author = {Nemirovsky, Daniel and Thiebaut, Nicolas and Xu, Ye and Gupta, Abhishek}, booktitle = {Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence}, pages = {1488--1497}, year = {2022}, editor = {Cussens, James and Zhang, Kun}, volume = {180}, series = {Proceedings of Machine Learning Research}, month = {01--05 Aug}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v180/nemirovsky22a/nemirovsky22a.pdf}, url = {https://proceedings.mlr.press/v180/nemirovsky22a.html}, abstract = {Model interpretability, fairness, and recourse for end users have increased as machine learning models have become increasingly popular in areas including criminal justice, finance, healthcare, and job marketplaces. This work presents a novel method of addressing these issues by producing meaningful counterfactuals that are aimed at providing recourse to users and highlighting potential model biases. A meaningful counterfactual is a reasonable alternative scenario that illustrates how input data perturbations can influence the model’s output. The CounteRGAN method generates meaningful counterfactuals for a target classifier by utilizing a novel Residual Generative Adversarial Network (RGAN). We compare our method against leading state-of-the-art approaches on image and tabular datasets over a variety of performance metrics. The results indicate a significant improvement over existing techniques in combined metric performance, with a latency reduction of 2 to 7 orders of magnitude which enables providing real-time recourse to users. The code for reproducibility can be found here: https://github.com/gan-counterfactuals/countergan.} }
Endnote
%0 Conference Paper %T CounteRGAN: Generating counterfactuals for real-time recourse and interpretability using residual GANs %A Daniel Nemirovsky %A Nicolas Thiebaut %A Ye Xu %A Abhishek Gupta %B Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2022 %E James Cussens %E Kun Zhang %F pmlr-v180-nemirovsky22a %I PMLR %P 1488--1497 %U https://proceedings.mlr.press/v180/nemirovsky22a.html %V 180 %X Model interpretability, fairness, and recourse for end users have increased as machine learning models have become increasingly popular in areas including criminal justice, finance, healthcare, and job marketplaces. This work presents a novel method of addressing these issues by producing meaningful counterfactuals that are aimed at providing recourse to users and highlighting potential model biases. A meaningful counterfactual is a reasonable alternative scenario that illustrates how input data perturbations can influence the model’s output. The CounteRGAN method generates meaningful counterfactuals for a target classifier by utilizing a novel Residual Generative Adversarial Network (RGAN). We compare our method against leading state-of-the-art approaches on image and tabular datasets over a variety of performance metrics. The results indicate a significant improvement over existing techniques in combined metric performance, with a latency reduction of 2 to 7 orders of magnitude which enables providing real-time recourse to users. The code for reproducibility can be found here: https://github.com/gan-counterfactuals/countergan.
APA
Nemirovsky, D., Thiebaut, N., Xu, Y. & Gupta, A.. (2022). CounteRGAN: Generating counterfactuals for real-time recourse and interpretability using residual GANs. Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 180:1488-1497 Available from https://proceedings.mlr.press/v180/nemirovsky22a.html.

Related Material