A Privacy-Preserving Federated Framework with Hybrid Quantum-Enhanced Learning for Financial Fraud Detection

ArXiv ID: 2507.22908 “View on arXiv”

Authors: Abhishek Sawaika, Swetang Krishna, Tushar Tomar, Durga Pritam Suggisetti, Aditi Lal, Tanmaya Shrivastav, Nouhaila Innan, Muhammad Shafique

Abstract

Rapid growth of digital transactions has led to a surge in fraudulent activities, challenging traditional detection methods in the financial sector. To tackle this problem, we introduce a specialised federated learning framework that uniquely combines a quantum-enhanced Long Short-Term Memory (LSTM) model with advanced privacy preserving techniques. By integrating quantum layers into the LSTM architecture, our approach adeptly captures complex cross-transactional patters, resulting in an approximate 5% performance improvement across key evaluation metrics compared to conventional models. Central to our framework is “FedRansel”, a novel method designed to defend against poisoning and inference attacks, thereby reducing model degradation and inference accuracy by 4-8%, compared to standard differential privacy mechanisms. This pseudo-centralised setup with a Quantum LSTM model, enhances fraud detection accuracy and reinforces the security and confidentiality of sensitive financial data.

Keywords: Federated Learning, Quantum LSTM, Fraud Detection, Privacy Preserving, Poisoning Attacks, Payments / Banking

Complexity vs Empirical Score

  • Math Complexity: 8.5/10
  • Empirical Rigor: 5.0/10
  • Quadrant: Holy Grail
  • Why: The paper employs advanced mathematical concepts, including quantum machine learning layers and differential privacy formulations, indicating high mathematical complexity. While the abstract presents specific quantitative results (e.g., 5% improvement, 4-8% reduction in degradation), the excerpt lacks detailed implementation code or backtest specifics, placing empirical rigor in the moderate range.
  flowchart TD
    A["Research Goal: Enhance financial fraud detection with privacy-preserving federated learning and quantum AI."] --> B["Methodology: Hybrid Quantum-Enhanced LSTM in a Federated Framework"]
    B --> C["Input: Distributed Financial Transaction Data"]
    C --> D["Process: Quantum LSTM Feature Extraction & Training"]
    D --> E["Process: FedRansel Privacy Preservation<br>(Defense vs. Poisoning/Inference Attacks)"]
    E --> F{"Outcome: Evaluation Metrics"}
    F --> G["Result 1: ~5% Improvement in Fraud Detection Accuracy"]
    F --> H["Result 2: 4-8% Reduction in Model Degradation/Inference Risk"]