Susanne Dandl, Christoph Molnar, Martin Binder, Bernd Bischl
TL;DR该研究提出了一种多目标反事实解释方法(MOC),将反事实搜索转化为多目标优化问题,通过对特征空间的多样性维护,返回一组不同权衡的反事实,并在实际案例中展示了 MOC 的有用性和与现有方法的对比。
Abstract
counterfactual explanations are one of the most popular methods to make
predictions of black box machine learning models interpretable by providing
explanations in the form of `what-if scenarios'. Most current ap