【Seminar】"A Bayesian approach to optimal quantum measurement - Magnetometry and POVMs"

Speaker

Julian Greentree, University of Melbourne

Title

"A Bayesian approach to optimal quantum measurement - Magnetometry and POVMs"

Abstract

Measurement is a central problem within the quantum information space, with implications for communications, computing, and fundamental theory. Optimal measurement requires three separate pillars: an accurate description of the system's dynamics; a complete method for interpreting and combining results; and an intelligent method to turn those results into new decisions. The first pillar is given by quantum mechanics, the second pillar by Bayes rule, and in this talk, Julian Greentree (PhD Candidate, University of Melbourne) presents an adaptive entropy reduction approach as candidate for the third pillar. Two applications will be discussed, adaptive magnetometry and sequential POVM design. In the context of magnetometry, entropy based adaptive control can be used to recreate an existing optimal algorithm presented by Kitaev. In the context of POVM design, adaptive control provides a generalised framework by which measurement procedures can be modelled, and how information theory and Bayesian techniques can be used to optimise them.

Keywords: Bayesian Probability, Adaptive Control, Shannon Entropy, Mutual Information, Adaptive Magnetometry, POVM design