From Paris to Yale, scientific collaboration at the heart of quantum physics

After two sabbaticals with Michel Devoret's team in 2011 and 2012, I divided my time between Inria and Yale University until the end of 2019. During this period, we collaborated on a wide range of topics related to the measurement, control, and stabilization of quantum systems, error correction, and fault tolerance in superconducting qubits, which resulted in some thirty theoretical and experimental articles.

With some of my students, we provided theoretical support for the experiments conducted by Michel Devoret's team, as well as by Robert Schoelkopf, his close collaborator at Yale. This stay in an experimental laboratory, marked by daily exchanges with Michel and the members of his team, enabled me to better understand the technical limitations and thus develop robust theoretical approaches to these constraints. As a result, many of the protocols we designed were quickly tested thanks to the experiments conducted by Michel Devoret and Robert Schoelkopf's teams, leading to major results published in journals such as Science and Nature. 

Among the projects carried out during this collaboration, the development of a new method for encoding quantum information in so-called Schrödinger cat states of the microwave electromagnetic field is undoubtedly the most significant result. In a series of theoretical and experimental articles beginning in 2013, we proposed exploiting the infinite-dimensional Hilbert space of a simple quantum system, a quantum harmonic oscillator, to encode the information of a quantum bit (qubit) while retaining the ability to correct certain types of errors. Furthermore, by equipping this system with a carefully designed nonlinear dissipative mechanism, we proposed to correct these errors autonomously.

These proposals led to the concept of cat qubits, which is now being studied by several academic groups and has been adopted as the preferred path toward a fault-tolerant quantum processor by companies such as Alice&Bob in France and Amazon Web Services in the United States.

Quantum error correction remains a key concern for most players in the field, with a view to building a fault-tolerant processor. The significant additional hardware costs involved are a major obstacle to the rapid development of reliable and truly useful machines. Theorists are continuing their research to develop new methods of encoding, protecting, and manipulating quantum information, with the aim of reducing these additional costs. For their part, experimenters are managing, year after year, to implement increasingly complex systems. I therefore hope that, in the not too distant future, these advances will converge and make it possible to build a reliable quantum processor, enabling the first concrete algorithms to be tested.