This project is a collaboration with EXXA and QMLHEP.
Variational quantum circuits (VQCs) lie at the core of many near-term quantum algorithms, enabling hybrid quantum–classical optimization of parametrized quantum circuits. Efficient VQC architectures underpin applications such as quantum machine learning (QML) and variational eigensolvers. However, the design of VQC architectures remains a largely heuristic and problem-dependent process that often involves extensive trial and error. VQCs are also very unintuitive for human designers, as quantum phenomena such as entanglement and interference have no classical analogs. As circuit depth and the number of qubits increase, the combinatorial circuit space rapidly becomes intractable for manual exploration, motivating new forms of automation. By leveraging the reasoning capabilities of Large Language Models (LLMs), this project aims to automate the synthesis and optimization of quantum circuits. Beyond initial design, the project implements an iterative “closed-loop” testing framework where the LLM interacts with quantum simulators and hardware backends to debug circuits, minimize gate depth, and verify logical equivalence. This approach seeks to bypass the “bottleneck” of manual circuit transpilation, potentially uncovering novel gate sequences that human intuition might overlook.
Total project length: 175/350 hours.
Intermediate/Advanced
Please DO NOT contact mentors directly by email. Instead, please email ml4-sci@cern.ch with Project Title and include your CV. The mentors will then get in touch with you.