In-network machine learning for real-time transaction fraud detection

Conference item

Machine learning (ML) has become a mainstream approach in the fight against transaction fraud for its intelligence. For financial institutions and businesses, low-latency detection of fraudulent transactions in real-time is highly important as it enables rapid identification and prevention. Concurrently mitigating fraudulent transactions by using ML while also reducing latency remains a challenging endeavor, for which performing inference within programmable network devices offers a potential solution. In this paper, we introduce MIND, conducting ML-based fraud detection within programmable devices. MIND is prototyped on both software and hardware network devices, including BMv2, Intel Tofino, and NVIDIA BlueField-2 DPU, and is evaluated with three publicly available transaction datasets. Experimental results demonstrate that MIND detects transaction fraud in real-time, with a throughput of 6.4 terabits per second and microsecond-scale latency. Compared with server-based solutions, MIND can process over ×800 more transactions per second, along with a latency reduction of over ×1300 per transaction. At the same time, MIND attains 99.94% of serverbased benchmarks’ accuracy and 93.66% of their F1-score, exhibiting only marginal degradation in classification performance. Therefore, MIND offers substantial savings in the number of servers, leading to reduced costs and energy consumption, while providing a better customer experience.

Authors: Hong, X; Zheng, C; Zilberman, N

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: European Association for Artificial Intelligence
• Host title: 27th European Conference on Artificial Intelligence, 19–24 October 2024, Santiago de Compostela, Spain – Including 13th Conference on Prestigious Applications of Intelligent Systems (PAIS 2024)
• Journal: Frontiers in Artificial Intelligence and Applications More from this journal
• Series: Frontiers in Artificial Intelligence and Applications
• Series number: 392
• Publication date: 2024-12-31
• Acceptance date: 2024-07-04
• Event title: 27th European Conference on Artificial Intelligence
• Event location: Santiago de Compostela
• Event website: https://www.ecai2024.eu/
• Event start date: 2024-10-19
• Event end date: 2024-10-24
• DOI: 10.3233/FAIA240828
• ISSN: 0922-6389
• ISBN: 9781643685489
• Language: English
• Keywords: Computer science; Database transaction; Database