Hybrid Quantum Risk Minimisation: A QRU-QAOA Pipeline for Spatial Flood Tail-Risk Allocation
IEEE Quantum Week 2026 — Quantum End-to-End Hybrid Case Studies (QECS) Technical Papers
Quantum Machine Learning · Data Streams
PhD candidate & freelance AI/ML consultant
I develop compact quantum and machine-learning systems for temporal data, from QRU theory and forecasting to explainability and downstream risk decisions.
About
I am Léa Cassé, a PhD candidate in a joint programme between the University of Waikato and École Polytechnique / Institut Polytechnique de Paris. My research focuses on Quantum Machine Learning for data streams, particularly the theory and practical design of Quantum Re-Uploading Units (QRUs) for time-series forecasting and downstream decision problems.
Alongside my PhD, I work as a freelance AI/ML consultant on governed machine-learning systems, semantic data contracts, forecasting evaluation, provenance, explainability, and human-in-the-loop review.
Research focus
Expressivity, trainability, Fourier structure, and architecture design for shallow QRU circuits under NISQ constraints.
Research 02Temporal modelling under drift, cross-day EMG classification, and attribution analyses that distinguish meaningful signal from unstable artefacts.
Projects 03QRU-to-QAOA pipelines, CVaR allocation, and decision-making under uncertainty for climate-risk applications.
ProjectsLatest news
Hybrid Quantum Risk Minimisation: A QRU-QAOA Pipeline for Spatial Flood Tail-Risk Allocation was accepted as a QCE26 technical paper in the Quantum End-to-End Hybrid Case Studies (QECS) track.
I will present the accepted work at IEEE Quantum Week 2026 in Toronto, Canada.
Selected work
IEEE Quantum Week 2026 — Quantum End-to-End Hybrid Case Studies (QECS) Technical Papers
2025 IEEE International Conference on Quantum Computing and Engineering (QCE25 / IEEE Quantum Week 2025)
arXiv:2412.12397