Pareto-optimal quantum kernel selection for unsupervised anomaly detection on real malware beaconing data

Key points

• Quantum kernel methods are leading candidates for a practical quantum advantage in machine learning, but assessing that potential requires two quantities usually reported separately: how well a kernel performs on the task, and how far its geometry departs from the classical kernels available for the same problem.1
• We apply it to malware beaconing detection in real network traffic, using a one-class support vector machine with fidelity and projected quantum kernels over four data encodings, on simulators and on IQM's 20-qubit Garnet processor.1
• NPD-guided selection alone finds a fidelity kernel that beats the tuned classical baseline, but with a geometric difference too small to certify the gain as quantum.1
• Projected kernels reach far larger geometric differences; the Pareto-selected one only marginally exceeds the baseline (AUC $0.782$ versus $0.765$, $g{C\to Q}\approx 89>\sqrt{N}$ relative to that reference kernel), still below the NPD-selected fidelity kernel ($0.840$).1

Sources (1)

• [1]Pareto-optimal quantum kernel selection for unsupervised anomaly detection on real malware beaconing data
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 09:13 AM

  We introduce a fully unsupervised, multi-objective protocol that optimises simultaneously the normalised pseudo discrepancy (NPD), a label-free proxy for anomaly detection quality, and the geometric difference (GD) to a tuned classical reference kernel, selecting models from the resulting Pareto front.

  Quantum kernel methods are leading candidates for a practical quantum advantage in machine learning, but assessing that potential requires two quantities usually reported separately: how well a kernel performs on the task, and how far its geometry departs from the classical kernels available for the same problem.

Extractive summary: sentences quoted from the sources.