“At the core of the project is the creation of a digital twin of a superconducting quantum processor, a virtual model that learns how the real processor behaves,” says Giovanna Tancredi, who leads one of two quantum technology research projects at Chalmers selected to receive support through the Wallenberg Foundations’ newly launched AI initiative.
A total of eleven research projects has been selected in the first round of the Knut and Alice Wallenberg AI Consultancy Grant, a new programme designed to help researchers rapidly integrate artificial intelligence into their research. Through the initiative, research teams gain access to AI expertise that can support new ways of working and accelerate scientific breakthroughs.
The grant scheme was launched in response to the rapid advances in artificial intelligence. Rather than providing traditional research funding, the programme gives researchers access to specialised AI expertise through AI4S, a company whose consultants combine advanced AI competence with research experience and work directly alongside the selected research groups.
Among the eleven projects awarded up to 700 hours of AI support are two projects from Chalmers, both based at the Department of Microtechnology and Nanoscience.
One of them is led by Giovanna Tancredi, Senior Researcher in Quantum Technology and leader of the quantum computer development effort within the Wallenberg Centre for Quantum Technology (WACQT). In the project quantum gate optimization with digital twins of superconducting qubits, submitted together with Research Specialist Tangyou Huang, they will use the AI support to develop a digital twin, a virtual model that learns how a real quantum processor operates. The model will be used to predict, analyze and optimize the processor’s performance.
“In an ideal quantum processor, control operation would work perfectly. In real devices, this is not true as they are affected by noise, unwanted interactions between qubits, changes in device conditions, and imperfections in control signals. These errors become increasingly difficult to understand as quantum processors grow larger. This project aims to develop a new approach for improving quantum processors by combining artificial intelligence (AI) with physics-based modelling,” says Giovanna Tancredi.
The second Chalmers project, AI-assisted calibration of superconducting quantum processors, is led by Göran Johansson, Professor of Applied Quantum Physics and Director of WACQT. The proposal was submitted by Michele Faucci Gianelli, who leads the auto-calibration activities within Chalmers’ quantum computing programme.
"This gives Michele Faucci Gianelli and the auto-calibration team the possibility to exploit state-of-the-art AI solutions to assist the tune-up of our 25-qubit quantum computer, which in turn will facilitate scaling towards 100 qubits," says Göran Johansson.
Artificial intelligence has rapidly become an increasingly important tool in research. AI can analyse vast amounts of data, identify patterns in complex systems and help researchers develop new models and hypotheses. The eleven selected projects were evaluated based on scientific excellence, the expected impact of AI on research, access to relevant data and the research teams’ commitment to integrating AI into their work.
“We are in the midst of a fundamental transformation in how research is conducted. AI creates unprecedented opportunities, as well as challenges, for academia. In many scientific fields, AI is opening the door to entirely new ways of working. However, Swedish researchers need access to advanced expertise and hands-on support to take full advantage of AI and address the challenges it presents,” said Anders Ynnerman, Deputy Executive Director of Knut and Alice Wallenberg Foundation, when the selected projects were announced.
The following projects and research groups have been awarded AI expertise:
• AI-assisted assignment from super-resolution NMR spectra, Vladislav Orekhov, University of Gothenburg
• AI-assisted calibration of superconducting quantum processors, Göran Johansson, Chalmers University of Technology
• AI-based feature learning from 5PSeq for AMR phenotyping, Vicent Pelechano, Karolinska Institutet
• AI-driven discovery of spatial host microbiota niches controlling mucosal healing in IBD, Eduardo Villablanca, Karolinska Institutet
• Domain-adaptive machine learning to overcome chemical domain shift between DNA-encoded libraries and billion-scale commercial chemical space, Jens Carlsson, Uppsala University
• Machine Learning Driven Modelling of Bio- and Conducting Polymers, Igor Zozoulenko, Linköping University
• Mining synthetic DNA selection data for hidden binders or targeting agents, Erik Benson, Karolinska Institutet
• Optimizing epilepsy treatment through population health data, Johan Zelano, University of Gothenburg
• Profit versus nature? Disentangling the polarized media discourse on forest management in Sweden, Leona Achtenhagen, Jönköping University
• Chalmers University of Technology, Quantum Gate Optimization with Digital Twins of Superconducting Qubits, Giovanna Sammarco Tancredi
• Spatial AI for Predicting Cell Fate Transitions and Disease Course in Multiple Sclerosis, Gonçalo Castelo-Branco, Karolinska Institutet
- Senior Researcher, Quantum Technology, Microtechnology and Nanoscience
- Full Professor, Applied Quantum Physics, Microtechnology and Nanoscience