Back
Solving Inverse Problems with Quantum Physics-Informed Machine Learning
The project aims at developing new quantum-accelerated computational methods for solving inverse problems in real-time, building upon recent developments in classical computational engineering. By implementing these methods on emerging real-world quantum computers, we can, e.g., potentially improve the efficiency and safety of future cars, including meeting the requirements for Level 5 autonomous driving. We explore the potential of physics-informed machine learning methods on quantum computers, focusing on demonstrating their effectiveness in solving inverse problems and optimizing complex systems. Last but not least, we investigate the impact of equation complexity on the real-time solvability, accuracy, and efficiency of physics-informed machine learning With this ambitious project, we hope to open new avenues for quantum computing.
Field of action:
Quantum Computing > Algorithms > Machine Learning
- Faculty 4 – Mechanical Engineering
- Modeling & Simulation Sciences (MSS, vormals CompSE / Jara CSD)
Address:
Lehrstuhl für Computergestützte Analyse Technischer Systeme (CATS), Schinkelstraße 2, 52062 Aachen
Contact:
Dr. Norbert Hosters
hosters@cats.rwth-aachen.de
Homepage:
http://www.cats.rwth-aachen.de
Status:
running