Detecting Potential Local Adversarial Examples for Human-Interpretable Defense
This tackles fraud prevention in industry applications like credit approval, but it appears incremental as an ongoing work with a first proposition.
The paper addresses the problem of detecting potential local adversarial examples in tabular data used for decisions like credit insurance approval, by identifying locally critical features to help human experts control information and prevent fraud.
Machine learning models are increasingly used in the industry to make decisions such as credit insurance approval. Some people may be tempted to manipulate specific variables, such as the age or the salary, in order to get better chances of approval. In this ongoing work, we propose to discuss, with a first proposition, the issue of detecting a potential local adversarial example on classical tabular data by providing to a human expert the locally critical features for the classifier's decision, in order to control the provided information and avoid a fraud.