Speaker
Sebastian Knauer
(University of Vienna)
Description
Developing novel quantum technology exhibits the challenge of their efficient characterisation. We introduce and experimentally demonstrate a methodology to automatically formulate and select Hamiltonian models, learning the most appropriate in reproducing the observed system’s dynamics. Here, we propose and experimentally demonstrate the quantum model learning agent (QMLA), a Bayesian approach based upon the generation and exploration of alternative, parametrised models; and additional a frequentist approach. To test our methodology, we use the Hamiltonian describing a nitrogen-vacancy-centre electron spin interacting with a spin bath.
Authors
Sebastian Knauer
(University of Vienna)
Raffaele Santagati
(Quantum Engineering Technology Labs, University of Bristol)
Antonio A. Gentile
(Quantum Engineering Technology Labs, University of Bristol)
Brian Flynn
(Quantum Engineering Technology Labs, University of Bristol)
Co-authors
Nathan Wiebe
(University of Washington)
Anthony Laing
(Quantum Engineering Technology Labs, University of Bristol)
Christopher E. Granade
(Quantum Architectures and Computation Group, Microsoft Research)
Stefano Paesani
(ntum Engineering Technology Labs, University of Bristol)
John G. Rarity
(Quantum Engineering Technology Labs, University of Bristol)
