Bayesian Quantum Neural Networks for Anomaly Detection

Bayesian quantum orthogonal neural networks for fault segmentation in 3D car parts and satellite ship-route imagery

Joint work with Airbus.

Applied Bayesian learning to Quantum Orthogonal Neural Networks for segmenting faults in 3D car parts, comparing feedforward and 3D convolutional architectures against their Bayesian counterparts to find the optimal solution, with real hardware experiments to benchmark performance (Mathur et al., 2025).

A related effort designed quantum autoencoders built from orthogonal neural networks to detect anomalies in satellite images of ship routes.

References

2025

  1. IEEE QCE
    Bayesian Quantum Orthogonal Neural Networks for Anomaly Detection
    Natansh Mathur, Brian Coyle, Nishant Jain, and 4 more authors
    In IEEE International Conference on Quantum Computing and Engineering (QCE), 2025
    HTML