Periodic Reporting for period 3 - AVaQus (ANNEALING-BASED VARIATIONAL QUANTUM PROCESSORS)
Periodo di rendicontazione: 2023-10-01 al 2024-09-30
The AVaQus project represents a launchpad towards developing a quantum computing technology based on quantum annealing with coherent superconducting qubits. The project defines coherent quantum annealers as those processors containing the following properties: being built out of qubits that retain coherence times much longer than the computation time; containing qubit-qubit interactions that have no classical analogue; the connectivity is high between qubit pairs; the annealer can be used beyond annealing problems, including hybrid quantum-classical variational problems following annealing-like processing techniques. The combination of all these properties results in a quantum device that has a high potential to overcome classical processors in the near term where no quantum error correction will be yet available. In addition, coherent quantum annealers can be scaled up in the long-run towards adiabatic quantum processors to perform universal quantum computation.
The goal of the AVaQus project is to develop a small-scale quantum annealer prototype with coherent flux-like qubits, with interactions beyond the Ising model. This prototype will be used as a demonstrator for several proof-of-principle dynamical control techniques and theoretically study its applicability in larger-scale processors that could lead to a quantum advantage. The annealer prototype will also be used to implement quantum algorithms beyond annealing such as quantum simulation and variational quantum eigensolvers. A final output of project AVaQus is a blueprint to build a large-scale coherent quantum annealing processor in a follow-up phase.
In concluding the project, a chain of three capacitively coupled flux qubits has been produced. The chain can be programmed as a two-qubit annealer with a tunable interaction that cannot be described by the usual Ising model. It can as well be programmed as a three-qubit system with fixed interactions for quantum simulation appliactions. In addition, enabling technology was produced along the project in the form of wideband tunable quantum limited amplifiers, as well as flexible cryogenic wiring that enhances the density of lines connecting to the quantum processors. Multiple algorithmic applications have been proposed such asusing annealers as optimizers with QAOA-type variational algorithms, as quantum simulators of spin models and as coupled fermions.
The goal of the AVaQus project is to develop a small-scale quantum annealer prototype with coherent flux-like qubits, with interactions beyond the Ising model. This prototype will be used as a demonstrator for several proof-of-principle dynamical control techniques and theoretically study its applicability in larger-scale processors that could lead to a quantum advantage. The annealer prototype will also be used to implement quantum algorithms beyond annealing such as quantum simulation and variational quantum eigensolvers. A final output of project AVaQus is a blueprint to build a large-scale coherent quantum annealing processor in a follow-up phase.
In concluding the project, a chain of three capacitively coupled flux qubits has been produced. The chain can be programmed as a two-qubit annealer with a tunable interaction that cannot be described by the usual Ising model. It can as well be programmed as a three-qubit system with fixed interactions for quantum simulation appliactions. In addition, enabling technology was produced along the project in the form of wideband tunable quantum limited amplifiers, as well as flexible cryogenic wiring that enhances the density of lines connecting to the quantum processors. Multiple algorithmic applications have been proposed such asusing annealers as optimizers with QAOA-type variational algorithms, as quantum simulators of spin models and as coupled fermions.
On the hardware side, multiple circuit components have been established as part of the final multi-qubit prototype. Among the completed components, a wideband nearly quantum-limited amplifier was produced with a bandwidth spanning several gigaherz and a center frequency which can be tuned over several gigaherz. At the same time, a flexible, scalable cabling technology has been adapted for quantum annealing needs and is ready to be tested with the multi-qubit devices. This cabling contains a high density of lines while exhibiting very low thermal conductivity, making it an ideal component for experiments involving large numbers of qubits and control channels, as is the case of annealers.
Individual coherent flux qubits of low and high impedance were built and characterized. Both types of qubits exhibit long coherence times in the microsecond range, making them suitable to build coherent quantum annealers at the few-qubit scale. As part of the production of low-impedance qubits, a new fabrication method employing niobium has been demonstrated, which may lead to flux qubits made with a larger-gap superconductor leading to lower losses.
A chain of three coupled high-impedance qubits was fabricated, using contactless flip-chip technology that enables replacing single qubit chips to select optimal devices. With such a system a tunably coupled qubit pair was demonstrated, using capacitive interactions which represent a model that goes beyond the standard Ising Hamiltonians and is necessary to overcome classical computational methods.
On the theory side, a circuit design generating a nonclassical interaction between flux-like qubits was produced using capacitive interactions. On the software side, a classical simulator was developed which allows the simulation of noisy quantum systems with reasonable accuracy for qubit numbers well above one hundred. In parallel, a variety of algorithms and simulation protocols have been developed, including a sorting algorithm, a method to engineer thermal states with a quantum annealer, and the simulation of interacting fermions.
The results of the project have been disseminated throughout the scientific community in important events worldwide, as well as in publications of high-impact journals such as Nature Physics and Physical Review Letters. All information regarding project progress and production has been updated on the project website avaqus.eu. Some of the results have been protected in the form of patents, with a total of 5 patents produced within the reporting period, 4 of which have already been published. It is expected that at least one more patent will be produced coming out of the results of the project.
Individual coherent flux qubits of low and high impedance were built and characterized. Both types of qubits exhibit long coherence times in the microsecond range, making them suitable to build coherent quantum annealers at the few-qubit scale. As part of the production of low-impedance qubits, a new fabrication method employing niobium has been demonstrated, which may lead to flux qubits made with a larger-gap superconductor leading to lower losses.
A chain of three coupled high-impedance qubits was fabricated, using contactless flip-chip technology that enables replacing single qubit chips to select optimal devices. With such a system a tunably coupled qubit pair was demonstrated, using capacitive interactions which represent a model that goes beyond the standard Ising Hamiltonians and is necessary to overcome classical computational methods.
On the theory side, a circuit design generating a nonclassical interaction between flux-like qubits was produced using capacitive interactions. On the software side, a classical simulator was developed which allows the simulation of noisy quantum systems with reasonable accuracy for qubit numbers well above one hundred. In parallel, a variety of algorithms and simulation protocols have been developed, including a sorting algorithm, a method to engineer thermal states with a quantum annealer, and the simulation of interacting fermions.
The results of the project have been disseminated throughout the scientific community in important events worldwide, as well as in publications of high-impact journals such as Nature Physics and Physical Review Letters. All information regarding project progress and production has been updated on the project website avaqus.eu. Some of the results have been protected in the form of patents, with a total of 5 patents produced within the reporting period, 4 of which have already been published. It is expected that at least one more patent will be produced coming out of the results of the project.
The main goal of project AVaQus is to produce small-scale coherent quantum annealing prototypes, of both low and high impedance, where anealing and simulation algorithms specifically designed can be executed. After conclusion of the project, the following developments have surpassed the existing state-of-the-art in their respective areas:
-Production of a wideband quantum-limited parametric amplifier with several gigaherz bandwidth, with center frequency tunable by several gigaherz, without exhibiting limitations in the amplification spectrum due to the pump. Such a device is superior than existing commercially available products, and compared to other experimental developments in the field. A key advantage is the exclusive use of aluminum and Josephson junctions to produce the device, compared to alternatives made of niobium. This technology is already available through the spin-off Silent Waves, thus having an impact in the whole community and other fields such as radioastronomy or radar.
-Development of flexible cryogenic wiring. This technology surpasses existing coaxial cabling in density of cables, increasing maximum number of cables dilution refrigerators from hundreds to thousands. The technology is commercially available an impacts all research and engineering connected with flexible wiring, both cryogenic and room temperature.
-Demonstration of the longest chain of coherent flux qubits on a scalable architecture. With the three superconducting flux qubits the device can perform as both quantum simulator as well as a quantum annealer. It can also be operated with gates, as the system exhibits long coherence times. The production of this prototype has an impact in the industry of quantum computing, in particular quantum annealing, as it paves the way to implement algorithms in this novel quantum computing platform. This prototype leads the path to obtaining a functional quantum processor with the capability to outperform conventional computers in the near term, with a strong societal impact in all the real-world problems it will be able to address. Annealers are particularly suited to optimization problems, which are extremely ubiquitous in our society, such as navigation, scheduling, and portfolio management, among many other relevant areas.
-Production of a wideband quantum-limited parametric amplifier with several gigaherz bandwidth, with center frequency tunable by several gigaherz, without exhibiting limitations in the amplification spectrum due to the pump. Such a device is superior than existing commercially available products, and compared to other experimental developments in the field. A key advantage is the exclusive use of aluminum and Josephson junctions to produce the device, compared to alternatives made of niobium. This technology is already available through the spin-off Silent Waves, thus having an impact in the whole community and other fields such as radioastronomy or radar.
-Development of flexible cryogenic wiring. This technology surpasses existing coaxial cabling in density of cables, increasing maximum number of cables dilution refrigerators from hundreds to thousands. The technology is commercially available an impacts all research and engineering connected with flexible wiring, both cryogenic and room temperature.
-Demonstration of the longest chain of coherent flux qubits on a scalable architecture. With the three superconducting flux qubits the device can perform as both quantum simulator as well as a quantum annealer. It can also be operated with gates, as the system exhibits long coherence times. The production of this prototype has an impact in the industry of quantum computing, in particular quantum annealing, as it paves the way to implement algorithms in this novel quantum computing platform. This prototype leads the path to obtaining a functional quantum processor with the capability to outperform conventional computers in the near term, with a strong societal impact in all the real-world problems it will be able to address. Annealers are particularly suited to optimization problems, which are extremely ubiquitous in our society, such as navigation, scheduling, and portfolio management, among many other relevant areas.