Janas QMLMS
Quantum machine learning for material science
The project
The JANAS‑QMLMS project—financed by ICSC Spoke 8 under PNRR/NextGenEU—blends classical high-performance computing, machine learning, and quantum computing to tackle industrial challenges in material sciences using current NISQ hardware. The project focuses on computing ground‑state energies of small molecules using neutral‑atom quantum platforms, using both analog and digital quantum paradigms as well as classical and hybrid machine learning techniques. Its hybrid approach leverages the strengths of each technique—digital, analog, and ML—to achieve improved accuracy and efficiency in materials‑science simulations.
Results
The JANAS-QMLMS project delivered two key results: an improved analog algorithm for quantum adiabatic optimization, enhanced through sampled quantum diagonalization; and a hybrid quantum–machine learning algorithm enabling efficient computation of low-level properties of small molecules, advancing materials modeling on NISQ devices.
Publications
Coming soon
Partnership – eXact lab
eXact lab is an SME with solid experience in the field of high-performance computing and data management. The team has a strong scientific background and is able to bridge the gap between scientific research and industrial needs.