Towards Responsible AI for Financial Transactions

TL;DR

This paper emphasizes responsible AI principles in finance, focusing on explainability through SHAP and hybrid methods, and evaluates model robustness against evasion attacks.

Contribution

It introduces a novel explanation approach combining SHAP and text clustering with decision trees for financial transaction models.

Findings

Effective feature importance analysis using SHAP

Hybrid explanation method enhances interpretability

Model robustness tested against targeted evasion attacks

Abstract

The application of AI in finance is increasingly dependent on the principles of responsible AI. These principles - explainability, fairness, privacy, accountability, transparency and soundness form the basis for trust in future AI systems. In this study, we address the first principle by providing an explanation for a deep neural network that is trained on a mixture of numerical, categorical and textual inputs for financial transaction classification. The explanation is achieved through (1) a feature importance analysis using Shapley additive explanations (SHAP) and (2) a hybrid approach of text clustering and decision tree classifiers. We then test the robustness of the model by exposing it to a targeted evasion attack, leveraging the knowledge we gained about the model through the extracted explanation.