Archer Materials Advances Quantum Machine Learning Fraud Detection Project

Archer Materials (ASX: AXE) has completed the next stage of its quantum machine learning (QML) fraud detection project after successfully testing and benchmarking a quantum neural network (QNN) model on a publicly available financial fraud dataset.

The early-stage model performed equivalently to the best classical models used in the benchmark test when run on a qubit simulator.

Archer said the model generated only one false positive while correctly identifying 118 fraudulent transactions.

The company also successfully executed the model on IQM Garnet, a commercial superconducting quantum computer accessed through AWS Braket.

Dataset Preparation Completed

Archer began the current phase after completing dataset preparation in March and moving the project into QML simulations and benchmarking.

The QNN workflow was developed and evaluated using a public financial fraud dataset containing more than 280,000 transaction records.

Archer processed the dataset using dimensionality reduction and data balancing techniques to allow operation within current quantum computing constraints.

The project used a staged experimental framework covering qubit selection studies, feature map optimisation, benchmarking against classical machine learning approaches, and quantum noise analysis.

Minimal False Alerts Recorded

The simulator test environment showed the QNN model could identify fraud while keeping false alerts low—one of the biggest practical challenges in fraud detection that can increase costs and create a poor customer experience.

The selected model remained stable under simulated quantum noise at low levels and showed only minor performance degradation at moderate noise levels.

Performance declined materially at higher noise levels, giving Archer information on the hardware quality and noise tolerance likely to be needed for future practical QML applications.

The company said the work had identified a high performing quantum architecture, established a repeatable benchmarking framework and provided insights into scaling, deployment constraints and noise tolerance.

IQM Hardware Validation

Archer also tested the QNN model on IQM Garnet, a 20-qubit superconducting quantum computer available through AWS Braket.

The hardware validation detected 18 of 19 fraudulent transactions in the test set, demonstrating that the model could operate on commercial quantum hardware.

The real hardware experiment involved higher false positive rates than simulator testing, but Archer said the results provided valuable validation under current quantum computing conditions.

Chief executive officer Dr Simon Ruffell said the simulator and hardware results provided an important validation step.

“These results have demonstrated that QML approaches can deliver strong fraud detection performance while operating within the constraints of current quantum computing systems,” Dr Ruffell said.

Prototype Target

Archer’s research collaboration agreement with the Commonwealth Scientific and Industrial Research Organisation (CSIRO) remains crucial.

“The research collaboration agreement with the CSIRO forms part of Archer’s strategy to investigate practical applications of quantum computing technologies and support future commercialisation opportunities in data-intensive industries,” Dr Ruffell said.

“Fraud detection is a relevant use case for QML because banks and payment providers must analyse large volumes of transaction data quickly, while reducing both missed fraud and false alerts.”

Archer said further work would include larger datasets, additional classical benchmarks, repeated trials, and further hardware validation before any commercial deployment pathway could be assessed.

The company is targeting a full QML prototype by the end of this year.