Quantum process tomography is aimed at learning unknown quantum processes from data. It is key to basically all applications of the quantum technologies, to build trust in the functioning of devices. However, known schemes are not sample-optimal and may suffer from state preparation and measurement (SPAM) errors. In this work, we introduce a scheme for quantum process tomography that is optimal in any desirable fashion. It makes use of
It relies heavily on new proof tools that have become available in the mathematical compressed sensing literature. This work has been published in the Physical Review Letters and selected as an Editor's choice.
News from Sep 25, 2018