Fed-RD: Privacy-Preserving Federated Learning for Financial Crime Detection

TL;DR

Fed-RD is a new federated learning algorithm designed for financial data that ensures privacy through differential privacy and secure computation, achieving high accuracy with minimal privacy trade-offs.

Contribution

It introduces Fed-RD, a novel privacy-preserving federated learning method tailored for relational financial data, combining differential privacy and secure multiparty computation.

Findings

High model accuracy with increased privacy levels

Outperforms benchmark federated learning methods

Effective on synthetic realistic datasets

Abstract

We introduce Federated Learning for Relational Data (Fed-RD), a novel privacy-preserving federated learning algorithm specifically developed for financial transaction datasets partitioned vertically and horizontally across parties. Fed-RD strategically employs differential privacy and secure multiparty computation to guarantee the privacy of training data. We provide theoretical analysis of the end-to-end privacy of the training algorithm and present experimental results on realistic synthetic datasets. Our results demonstrate that Fed-RD achieves high model accuracy with minimal degradation as privacy increases, while consistently surpassing benchmark results.