Superconducting quantum circuits is one of the most promising approaches to building a quantum computer. A reason for this is the increase in the coherence and relaxation time of superconducting qubits over the last decades, starting from less than a nanosecond in the early 2000s to several hundreds of microseconds nowadays. Thanks to these advances, it is possible to run simple quantum algorithms with tens of these qubits. However, scaling up to larger quantum processors still requires a significant improvement in the quality of all its components. Therefore, the objective of this project is to design more robust superconducting qubits. Among the promising avenues are bosonic qubits, where quantum information is stored in high-quality electromagnetic cavities, and variations on the 0-π qubit which exploits symmetries to decouple the qubit from its noisy environment. This theoretical work will be done in close collaboration with experimental groups at partner institutions such as UC Berkeley and MIT from the Quantum System Accelerator, a US-funded quantum technology center.

Quantum processors based on superconducting qubits have already been used to reach quantum advantage, where the quantum processor outperforms classical processors at a specific task. Despite this remarkable achievement, current quantum processors are still too error-prone to perform useful computations. A major challenge in this field is to design useful quantum algorithms for today’s small-scale quantum computers. A class of quantum algorithms that is particularly promising is known as hybrid quantum algorithms, where a classical processor calls upon a quantum processor to solve an optimization task. To extract as much computational power as possible from the current, imperfect, quantum computers, we propose to adapt hybrid quantum algorithms to the specific architecture of superconducting quantum processors. A key step in this process will be to express hybrid quantum algorithms in terms of gates that are native to these processors. Moreover, to understand how to mitigate its effect, we propose to explore the effect of realistic noise on the performance of hybrid quantum algorithms. Tests on real devices with up to 65 qubits (as of today) will be performed using the IBM quantum processors available to the group through the Institut quantique’s IBM Q Hub.

Even though there is much evidence of its existence, the nature of dark matter remains one of the great mysteries of modern physics. Indeed, none of the known fundamental particles have the adequate properties to compose this non-conventional matter. A number of candidate particles have been proposed, among which are axions and dark photons, yet experimental research for the detection of these and other candidate dark matter particles has so far been unsuccessful. An important challenge is that the expected signal is very weak and is therefore drowned out by a significant background noise. New and more sensitive measurement techniques must therefore be developed, and quantum technologies based on superconducting quantum circuits are a particularly promising avenue. First experiments have already demonstrated that superconducting quantum circuits can be used to increase the sensitivity of axion detection and thus accelerate their discovery. The objective of the project is to develop more sensitive methods for the detection of dark matter axions using superconducting quantum circuits. As part of an international and multidisciplinary team, the candidate will work collaboratively with experimental groups at the Institut quantique and worldwide in order to accelerate our understanding of the nature of dark matter.