FRAUD-RLA: A new reinforcement learning adversarial attack against credit card fraud detection

TL;DR

This paper introduces FRAUD-RLA, a reinforcement learning-based adversarial attack that effectively bypasses credit card fraud detection systems with minimal prior knowledge, highlighting new vulnerabilities.

Contribution

The paper presents a novel reinforcement learning attack model specifically designed for credit card fraud detection, addressing limitations of existing attacks and demonstrating its effectiveness across multiple datasets.

Findings

FRAUD-RLA successfully bypasses fraud classifiers in experiments.

Effective with less prior knowledge than existing methods.

Works across diverse datasets and detection systems.

Abstract

Adversarial attacks pose a significant threat to data-driven systems, and researchers have spent considerable resources studying them. Despite its economic relevance, this trend largely overlooked the issue of credit card fraud detection. To address this gap, we propose a new threat model that demonstrates the limitations of existing attacks and highlights the necessity to investigate new approaches. We then design a new adversarial attack for credit card fraud detection, employing reinforcement learning to bypass classifiers. This attack, called FRAUD-RLA, is designed to maximize the attacker's reward by optimizing the exploration-exploitation tradeoff and working with significantly less required knowledge than competitors. Our experiments, conducted on three different heterogeneous datasets and against two fraud detection systems, indicate that FRAUD-RLA is effective, even considering the severe limitations imposed by our threat model.