Designing an attack-defense game: how to increase robustness of financial transaction models via a competition

TL;DR

This paper introduces a novel competition framework to evaluate and improve the robustness of financial transaction models against adversarial attacks, using a new dataset and real-world simulation of attack-defense dynamics.

Contribution

It presents a unique adversarial competition structure for financial transaction models, along with a new dataset, to study robustness in realistic scenarios.

Findings

Top submissions demonstrated varying levels of robustness.

The competition revealed key vulnerabilities in current models.

The new dataset enables practical research in financial security.

Abstract

Banks routinely use neural networks to make decisions. While these models offer higher accuracy, they are susceptible to adversarial attacks, a risk often overlooked in the context of event sequences, particularly sequences of financial transactions, as most works consider computer vision and NLP modalities. We propose a thorough approach to studying these risks: a novel type of competition that allows a realistic and detailed investigation of problems in financial transaction data. The participants directly oppose each other, proposing attacks and defenses -- so they are examined in close-to-real-life conditions. The paper outlines our unique competition structure with direct opposition of participants, presents results for several different top submissions, and analyzes the competition results. We also introduce a new open dataset featuring financial transactions with credit default labels, enhancing the scope for practical research and development.