DPxFin: Adaptive Differential Privacy for Anti-Money Laundering Detection via Reputation-Weighted Federated Learning

2026-03-14

AI Summary

This work addresses the challenge of anti-money laundering (AML) detection, which requires preserving the privacy of sensitive tabular data while capturing complex fraud patterns. Conventional federated learning approaches often fall short in preventing privacy leakage. To overcome this limitation, the authors propose DPxFin, a novel framework featuring a reputation-guided adaptive differential privacy mechanism. This mechanism dynamically evaluates each client’s local model consistency with the global model to quantify its reputation and accordingly allocates privacy-preserving noise: clients with high reputation receive less noise to enhance model utility, while those with low reputation are assigned stronger noise for improved privacy guarantees. Experimental results on both IID and non-IID AML tabular datasets using an MLP-based federated architecture demonstrate that DPxFin achieves a superior trade-off between accuracy and privacy, effectively mitigating tabular data leakage attacks.

Technology Category

• Machine Learning: Privacy
• Natural Language Processing: Ethics — Bias, Fairness, Transparency & Privacy
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Security and Privacy: Data transparency and provenance
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

In the modern financial system, combating money laundering is a critical challenge complicated by data privacy concerns and increasingly complex fraud transaction patterns. Although federated learning (FL) is a promising problem-solving approach as it allows institutions to train their models without sharing their data, it has the drawback of being prone to privacy leakage, specifically in tabular data forms like financial data. To address this, we propose DPxFin, a novel federated framework that integrates reputation-guided adaptive differential privacy. Our approach computes client reputation by evaluating the alignment between locally trained models and the global model. Based on this reputation, we dynamically assign differential privacy noise to client updates, enhancing privacy while maintaining overall model utility. Clients with higher reputations receive lower noise to amplify their trustworthy contributions, while low-reputation clients are allocated stronger noise to mitigate risk. We validate DPxFin on the Anti-Money Laundering (AML) dataset under both IID and non-IID settings using Multi Layer Perceptron (MLP). Experimental analysis established that our approach has a more desirable trade-off between accuracy and privacy than those of traditional FL and fixed-noise Differential Privacy (DP) baselines, where performance improvements were consistent, even though on a modest scale. Moreover, DPxFin does withstand tabular data leakage attacks, proving its effectiveness under real-world financial conditions.

Problem

• Anti-Money Laundering
• Differential Privacy
• Federated Learning
• Privacy Leakage
• Tabular Data

Innovation

• Adaptive Differential Privacy
• Reputation-Weighted Federated Learning
• Anti-Money Laundering
• Tabular Data Privacy
• Dynamic Noise Allocation