Users and their projects

Molecular electronic structure calculations
on the VLQ quantum computer

Call: 37th Open Access Grant Competition; OPEN-37-22

Researcher: Libor Veis

Institution: J. Heyrovský Institute of Physical Chemistry, Czech Academy of Sciences

Field: Material Sciences

Libor Veis has been awarded computing time on the VLQ quantum computer to develop and test new quantum-classical methods for molecular electronic structure calculations. On VLQ, the researchers will test variational quantum algorithms that combine quantum computing with classical methods. They will focus in particular on molecules with complex electronic bonds, such as systems important for studying chemical reactions or catalysts, where an accurate description of electronic structure is extremely challenging.

The aim is to assess the accuracy, scalability and robustness of these methods against the noise of current quantum computers and to determine which types of molecules they could benefit in the future. The research is part of the “Multiscale Chemical Simulations on Noisy Quantum Computers (MUSIQ)” project supported by the Czech Science Foundation and will contribute to the development of quantum computational chemistry and the exploration of new possibilities for using quantum technologies in scientific computing.

Quantum-enhanced anti-money laundering using dynamic quantum reservoir computing

Call: 37th Open Access Grant Competition; OPEN-37-1

Researcher: Van Binh Henri VU

Institution: IT4Innovations

Field: Informatics

Van Binh Henri VU has been awarded computing time on the Barbora NG, Karolina and LUMI supercomputers, as well as on the VLQ quantum computer, to develop a new method for detecting money laundering using quantum technologies and artificial intelligence. Banks already use artificial intelligence to identify suspicious patterns in financial transactions, but today's "classical" computers often struggle to keep pace with increasingly sophisticated and rapidly evolving methods used by fraudsters.

Van Binh Henri VU is therefore developing a type of brain-inspired quantum system - a hybrid quantum-classical reservoir designed to process company transaction time series, using a dataset and preprocessing pipeline adapted from recent published research on transaction-based fraud detection. Each transaction is fed one at a time into a small, fixed random quantum circuit; the resulting stream of measurements is combined into a reservoir "trajectory" that summarizes a company's transaction history, which a simple classifier then uses to flag suspected fraud.

On the VLQ quantum computer, he will test this approach - known as hybrid quantum-classical reservoir computing - and compare it against classical baselines, including a matched classical reservoir and standard sequence models such as LSTMs, RNNs and GRUs. The aim is to determine whether quantum technologies can enable more accurate detection of financial fraud and reduce the number of false alarms, and to understand how the underlying complexity of transaction data relates to detection performance.

The impact of noise on quantum computing
of the electronic structure of solids

Call: 37th Open Access Grant Competition; OPEN-37-12

Researcher: Martin Friák

Institution: Institute of Physics of Materials of the Czech Academy od Sciences

Field: Material Sciences

Martin Friák has been awarded computing time on the VLQ quantum computer to research the potential of quantum technologies for calculations of the electronic structure of crystals. Knowledge of electronic structure is essential for accurately predicting material properties, including their physical and chemical characteristics. However, for larger systems, calculations performed on classical computers face high computational demands and limited scalability.

Quantum computers offer a new approach, but current generations are still affected by noise and errors that can influence results. Researchers will therefore use the real VLQ quantum computer to investigate how these factors affect electronic structure calculations and compare the results with quantum simulations performed on the Karolina and Barbora NG supercomputers.

The goal is to better understand the capabilities of current quantum computers and to develop new algorithms and hybrid quantum-classical methods that can help reduce the impact of noise and errors.

Hybrid quantum-classical methods for robust audio deepfake detection

Call: 37th Open Access Grant Competition; OPEN-37-85

Researcher: Anton Firc

Institution: Brno University of Technology

Field: Informatics

Anton Firc has been awarded computing time on the Karolina and LUMI supercomputers and the VLQ quantum computer to develop new methods for detecting so-called audio deepfakes. Artificial intelligence can now generate increasingly realistic synthetic voices, creating challenges for the trustworthiness of digital communication, media, and security systems.

Researchers from Brno University of Technology are therefore developing hybrid models that combine classical artificial intelligence with quantum components. They will investigate whether these quantum components can improve the reliability of systems for detecting artificially generated speech and determine how they can be most effectively applied to audio data analysis.

The project will also focus on finding more efficient ways to represent and process speech data and on optimising computational workflows. The goal is to develop new approaches for audio forensic analysis that are more robust to changing conditions and could, in the future, contribute to better protection of digital communication.

Quantum and classical computing for Auger spectroscopy of molecules

Call: 37th Open Access Grant Competition; OPEN-37-15

Researcher: Maximilián Lamanec

Institution: IT4Innovations

Field: Material Sciences

Maximilián Lamanec has been awarded computing time on the Karolina, LUMI and Barbora NG supercomputers, as well as on the VLQ quantum computer, to develop new computational methods that will enable more accurate interpretation of experiments focused on molecular properties. The project focuses on Auger spectroscopy, a method that uses X-rays to study the electronic structure of molecules and processes related to their electronic states.

The project includes the development of a theoretical framework for more accurate simulations of Auger spectra, enabling a better link between experimental data and the electronic structure of molecules. On the VLQ quantum computer, Maximilián Lamanec will test quantum-classical computational approaches for calculating the electronic states of molecules and explore their potential for future use in demanding chemical simulations.

The research results could contribute to the development of methods applicable in chemistry, materials research and other fields where a detailed understanding of molecular properties is important.

High accuracy machine learning force fields
for pharmaceutical research

Call: 37th Open Access Grant Competition; OPEN-37-77

Researcher: Przemyslaw Karol Grenda

Institution:  Charles University

Field: Biosciences

Przemysław Grenda has been awarded computing time on the Karolina and LUMI supercomputers and the VLQ quantum computer to develop new methods combining artificial intelligence, quantum chemistry and quantum computing for simulations relevant to pharmaceutical research. Many modern medicines are processed in the human body by enzymes from the cytochrome P450 family, making an understanding of their behavior essential for the development of safe and effective drugs. At the center of every one of these enzymes sits a heme group, the same iron-containing structure that carries oxygen in hemoglobin and gives blood its red color. In cytochrome P450, however, the iron does something quite different: instead of simply binding oxygen, it uses it to attack and break apart drug molecules. It is precisely this chemistry that makes these enzymes so difficult to simulate.

The researchers are therefore developing machine learning models that can reproduce the results of accurate quantum-mechanical calculations at a significantly lower computational cost. The aim is to create a tool that enables faster, more accurate simulations of interactions between drugs and enzymes involved in their metabolism, which could help reduce the time and financial costs associated with developing new medicines.

The project will also explore using the VLQ quantum computer to generate data needed to train these models. On VLQ, the researchers will test methods for calculating atomic forces using quantum algorithms and investigate their potential use as a basis for developing more accurate machine learning models.

Quantum state tomography from incomplete set
of measurements

Call: 37th Open Access Grant Competition; OPEN-37-38

Researcher: Pavel Baláž

Institution: Institute of Physics of the Czech Academy of Sciences

Field: Material Sciences

Pavel Baláž has been awarded computing time on the Karolina supercomputer and the VLQ quantum computer to develop a new method that uses artificial intelligence to make it easier to determine the state of quantum systems. To use quantum computers effectively, researchers need to accurately identify the state of their qubits. This process is known as quantum tomography, but for larger quantum systems it requires a huge number of repeated measurements, making it extremely time-consuming and costly.

The project is part of the AI: Artificial Intelligence for Science and Society programme, one of the programmes within the AV21 Strategy. As part of the project, researchers from the Institute of Physics of the Czech Academy of Sciences are developing an artificial intelligence-based method that can reconstruct the most likely state of a quantum system from measurement results, which are inevitably affected by noise. One of the key questions addressed by the project is how many measurements are actually required: whether a complete set of measurements is necessary, or whether the model can work with a significantly smaller amount of data.

On the VLQ quantum computer, researchers will generate data from quantum circuits, which will be used to train and validate the new model. Using the Karolina supercomputer, they will then test its capabilities for different sizes of quantum systems and compare it with other artificial intelligence-based approaches. The goal is to develop a more efficient way of characterising quantum systems that can support the further development of quantum technologies.

Quantum interval power flow

Call: 37th Open Access Grant Competition; OPEN-37-13

Researcher: Michal Belina

Institution: IT4Innovations

Field: Engineering

Michal Belina has been awarded computing time on the Karolina supercomputer and the VLQ quantum computer to research the use of quantum computing for the analysis of power grids. The project focuses on interval power flow calculations, which help verify the safe operation of power networks even under uncertainty caused, for example, by variable energy production from renewable sources.

Michal Belina will investigate whether quantum algorithms can provide a more efficient way of identifying critical operating conditions than current methods. These methods gradually propagate uncertainty in input values through all calculation equations. As a result, errors accumulate and the resulting limits become wider than what actually reflects the real situation – making the grid operation appear riskier than it really is. The quantum approach avoids this issue by formulating the problem as an optimisation task and directly searching for the combination of inputs that leads to the worst-case operating condition. The result could be narrower limits for voltage and power flows, enabling more efficient use of existing infrastructure without compromising safety.

On the VLQ quantum computer, this approach will be tested on smaller power grid models and the results will be compared with classical methods. The goal is to evaluate the potential of quantum technologies for the future analysis of energy systems.

The research is carried out within the project “Increasing the resilience of energy grids in the context of decarbonisation, decentralisation and sustainable socio-economic development”, funded by the Johannes Amos Comenius Operational Programme.