Advanced qubit measurement for quantum computing, which consists of analyzing a system’s physical properties to generate a “space-time map” of the phenomena at work inside a quantum processing unit, is one of the pillars of Quobly’s R&D strategy. Measurement data will enable Quobly to characterize the behavior of individual qubits inside its chips and come up with effective approaches for dealing with noise. Quobly’s growing quantum information team counts among its members scientists involved in recent advances in measurement demonstrated on Google’s quantum processor.
Advanced qubit measurement protocols a pillar of Quobly’s R&D
A recent advance in quantum simulation on a 69-qubit superconducting processor at Google leveraged a qubit measurement protocol developed in part by Benoit Vermersch, who leads Quobly’s growing quantum information team—a team that also reflects the company’s full-stack approach to quantum, with Alain Champenois and Thibaud Louvet working on algorithms, and a partnership with CEA-Leti’s Valentin Savin for quantum error correction.
The results of the Google chip experiment, published in Nature in February 2025, highlighted the superior measurement capabilities of a hybrid analog-digital quantum simulator. The quantum processor at Google, with both analog and digital control, was used to simulate the thermalization processes, critical phenomena, and the interplay between quantum and classical behavior within a 2D XY magnet—going beyond what can be achieved using classical approaches.
Qubit entanglement dynamics vital to scaling quantum computing
The measurements used to characterize the system’s entanglement dynamics in this groundbreaking experiment included randomized measurement techniques co-developed by Quobly’s Vermersch to characterize entanglement scaling. The research that resulted in these measurements began when Vermersch was a postdoctoral researcher at the University of Innsbruck and, later, when he was Assistant Professor at Grenoble-Alpes University. In the case mentioned above, multiple measurements of physical qubit properties were used to provide objective information about the states within Google’s quantum chip.
Google’s quantum processor has made the news many times in recent months. What makes this demonstration different is the use of an analog mode where all qubits are operating together. As all couplings are “on” all the time, the system dynamics are no longer described by a quantum circuit, with digitally-controlled gates switched on and off, but by a many-body Hamiltonian. This enables quantum simulations where the Hamiltonian can be programmed to emulate and understand the rich physics of condensed matter systems. Quobly will use the same kind of randomized measurements used by Google to reveal the entanglement content and certify the high fidelity of our own quantum processors.
To fight noise in quantum systems, you must first measure it
The ability to measure physical properties to “see” inside arrays of qubits provides valuable information on qubit fidelity (i.e., the extent to which the state obtained during an experiment aligns with the ideal state). If a measurement reveals that an individual qubit is not behaving as intended, the code can be rewritten to exclude the qubit from future computations. Reliable data can also enable more sophisticated error mitigation approaches, which involve adapting the quantum circuit to the type of noise measured in an experiment. IBM, for example, has reported using this kind of approach.
Randomized qubit measurement part of Quobly’s full-stack approach to quantum
Quobly’s measurement toolbox is based on physical properties, which makes it hardware agnostic. And our team has developed an efficient theoretical framework and software code for gathering and processing the large volumes of bitstrings necessary for the effective certification of large quantum systems.
Measurement & benchmarking | To obtain objective data on observables, entanglement, fidelity, and other states. |
Numerical simulation & modeling | To predict how qubits will behave at scale. |
Quantum architectures & error correction | To devise strategies for implementing high-fidelity quantum circuits, with combined use of error mitigation and error correction. |
Fault-tolerant quantum algorithm R&D | To determine which quantum algorithms will work best on our hardware. |
Our quantum information tools are designed specifically for the challenges of large-scale quantum systems. Part of our strategy for developing a universal quantum computer involves contributing actively to the full-stack ecosystem, from error correction to algorithms. Along these lines, our qubit measurement R&D work, part of our broader R&D strategy to develop a universal quantum processor, will focus on:
- Adapting this measurement method for even larger quantum states.
- Revealing more physical properties with better accuracy.
- Taking advantage of new possibilities for measuring a quantum system using, for example, entangling measurements (the process of data acquisition that uses entangling gate operations).
For our own hardware development needs, experience demonstrating large-scale measurement protocols will enable us to implement immediately-useful demonstration experiments at Quobly and drive meaningful efforts toward developing a new generation of protocols tailored to the specificities of Quobly’s quantum hardware. We believe that in-house advanced qubit measurement for quantum computing will help speed up our progress toward 1,000 logical qubits and get us there faster than any other quantum company.
The science behind qubit measurement techniques
- Probabilistic error cancellation with sparse Pauli–Lindblad models on noisy quantum processors.
- Thermalization and criticality on an analogue–digital quantum simulator.
- Robust Estimation of the Quantum Fisher Information on a Quantum Processor.
- Observing the Quantum Mpemba Effect in Quantum Simulations. Selected for Physical Review Letters collection of the year 2024.
- The randomized measurement toolbox.
Author
