Quantum model improves anomaly detection with scarce data
Quantum anomaly detection in real scarce data
Machine Learning
Summary
Detecting unusual or rare events in small or unbalanced datasets is a tough task for traditional machine learning because anomalies are rare and unpredictable. The authors propose a new approach that combines classical and quantum computing to build a model with fewer parameters that works well on smaller datasets and is more interpretable. They tested their method on data from large solar power plants and found it can detect anomalies effectively and generalize well. This approach holds promise for making anomaly detection more efficient and sustainable using emerging quantum hardware.
What this means in practice
- •For energy plant operators: Detect unusual events automatically in large-scale solar power plants using a quantum-classical model trained on limited data.
- •For cybersecurity analysts: Improve detection of rare and diverse anomalies in scarce or unbalanced datasets for network security monitoring.
Authors
Emanuele Casciaro, Fabio Mascherpa, Alfonso Amendola, Filippo Caruso
Abstract
Anomaly detection on small and unbalanced datasets remains very challenging in machine learning, although this scenario is common in several domains, including healthcare, cybersecurity, finance, and energy. Data augmentation and generative AI may mitigate training-data scarcity, but they often fall short because anomalies are, by definition, unpredictable, rare, and highly diverse events compared to high-probability normal data. Overfitting to pseudo-anomalies, model collapse, high-dimensional data, uninterpretable black-box models, and validation challenges are typical issues limiting their practical applicability. In this context, quantum machine learning may provide a promising and more sustainable avenue because it can enable more interpretable models with far fewer trainable parameters and smaller datasets, implementable on energy-efficient quantum hardware. Here, we propose a novel two-step hybrid classical--quantum architecture for sequential data and test it on a realistic scenario in the global energy-transition domain, i.e., automated anomaly detection in large-scale photovoltaic plants. The achieved generalization capability and competitive prediction accuracy may pave the way for new hybrid learning models able to exploit the continuously increasing power of cloud-available and more sustainable quantum accelerators integrated with more traditional energy-hungry High Performance Computing resources.