CounteRGAN: Generating Realistic Counterfactuals with Residual Generative Adversarial Nets
Daniel Nemirovsky111Work done while at Hired. Amazon Seattle, WA U.S.A.、Nicolas Thiebaut Hired San Francisco, CA U.S.A.、Ye Xu* Facebook Menlo Park, CA U.S.A.、Abhishek Gupta* Facebook Menlo Park, CA U.S.A.
Abstract
The prevalence of machine learning models in various industries has led to growing demands for model interpretability and for the ability to provide meaningful recourse to users. For example, patients hoping to improve their diagnoses or loan applicants seeking to increase their chances of approval. Counterfactuals can help in this regard by identifying input perturbations that would result in more desirable prediction outcomes. Meaningful counterfactuals should be able to achieve the desired outcome, but also be realistic, actionable, and efficient to compute. Current approaches achieve desired outcomes with moderate actionability but are severely limited in terms of realism and latency. To tackle these limitations, we apply Generative Adversarial Nets (GANs) toward counterfactual search. We also introduce a novel Residual GAN (RGAN) that helps to improve counterfactual realism and actionability compared to regular GANs. The proposed CounteRGAN method utilizes an RGAN and a target classifier to produce counterfactuals capable of providing meaningful recourse. Evaluations on two popular datasets highlight how the CounteRGAN is able to overcome the limitations of existing methods, i
中文速览
机器学习模型在医疗诊断、贷款审批等高stakes场景中广泛应用，但当模型给出不利预测时，用户迫切需要知道"怎么改变现状"——这就是所谓的反事实（counterfactual）推荐，即找到对输入数据的最小调整，使预测结果朝有利方向转变。现有方法依赖梯度下降逐点搜索，不仅速度极慢，生成的反事实也往往脱离现实数据分布，缺乏可操作性。本文提出CounteRGAN方法，将生成对抗网络（GAN）引入反事实搜索，并设计了一种新颖的残差生成对抗网络（Residual GAN，RGAN）——让生成器直接学习输出扰动量而非完整数据点，从而有效缓解模式崩溃问题、提升反事实的真实性与可操作性。在两个主流数据集上的实验表明，CounteRGAN在保持预测改善效果和可操作性的同时，推理延迟比现有最优方法缩短了50倍至9万倍以上，真正实现了实时响应，为各行业ML系统向用户提供公平、透明、有实际意义的建议奠定了基础。
原文 arXiv:2009.05199;中英对照 + 大白话阅读 https://aha.fim.ai/paper/2009.05199v2