Mainz 2026 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
Q: Fachverband Quantenoptik und Photonik
Q 67: Poster – Quantum Information
Q 67.22: Poster
Donnerstag, 5. März 2026, 17:00–19:00, Philo 2. OG
Quantum Kolmogorov-Arnold Networks for Interpretable Healthcare Models — •Vanessa Stein1, Yannick Werner1,2, Akash Malemath1, Nikolaos Palaiodimopoulos1,2, Paula Manso Zorilla2, Hamraz Javaheri2, Mengxi Liu2, Paul Lukowicz1,2, Vitor Fortes Rey1,2, Gregor Alexander Stavrou3, Omid Ghamarnejad3, and Maximilian Kiefer-Emmanouilidis1,2 — 1RPTU Kaiserslautern-Landau — 2DFKI Kaiserslautern — 3Department of General, Visceral and Oncological Surgery, Klinikum Saarbrücken
Transparency remains a major challenge in applying machine learning to healthcare, where understanding model decisions is crucial for clinical trust and adoption. Kolmogorov-Arnold networks provide a structurally interpretable alternative to conventional neural architectures by placing learnable functions on edges rather than nodes. We explore quantum realizations of these architectures using Quantum Circuit Born Machines to create compact, expressive models with directly accessible functional components. Initial results indicate that the approach enables interpretable decision pathways without sacrificing predictive performance compared to established classical and quantum classifiers. This highlights the potential of quantum machine learning to support trustworthy AI in healthcare by combining strong performance with enhanced interpretability.
Keywords: Kolmogorov Arnold Network; Quantum Computing; Quantum Machine Learning; Health; AI