BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Anomalous Distillation of Metrological Quantum Information - David
  Arvidsson Shukur\, Hitachi Cambridge Laboratory
DTSTART:20220310T141500Z
DTEND:20220310T151500Z
UID:TALK170819@talks.cam.ac.uk
CONTACT:Damian Pitalua-Garcia
DESCRIPTION:Quantum experiments that involve post-selection or filtering h
 ave generated heated debate over the last decades— from a foundational a
 nd a practical perspective. Pre- and post-selected expectation values of a
  weakly measured observable can lie outside an observable’s eigenspectru
 m. Classical theories cannot explain this. In this talk\, I investigate po
 st-selection within the framework of single- and multi-parameter quantum l
 earning.\n\nQuantum learning (in metrology and machine learning) involves 
 estimating unknown parameters 𝜽 = (𝜃_1\, 𝜃_2\,… \, 𝜃_𝑀) f
 rom measurements of quantum states 𝜌_𝜽. The quantum Fisher informati
 on matrix can bound the average amount of information learnt about 𝜽 pe
 r experimental trial. In several scenarios\, it is advantageous to concent
 rate information in as few states as possible\, for example\, to avoid det
 ector saturation. Here\, we present a “go-go” theorem proving the poss
 ibility of unbounded and lossless distillation of Fisher information about
  multiple parameters in quantum learning. That is\, there is no cap on how
  much quantum information can be distilled into a subset of quantum partic
 les. My\, and collaborators’\, results enable the construction of filter
 s that can reduce arbitrarily the number of quantum states\, whilst retain
 ing all initial information. The fundamental resource underlying this unbo
 unded information distillability\, I will show\, is Kirkwood-Dirac negativ
 ity\, a narrower non-classicality concept than non-commutation.\n\nThe rec
 ipe for lossless and unbounded distillation of quantum information\, exten
 ds pre- and post-selected techniques of weak-value amplification to the gr
 owing fields of quantum machine learning and quantum metrology. If time pe
 rmits\, I will describe a proof-of-principle experiment where collaborator
 s and I demonstrate how the use of quantum-filtering can boost the informa
 tion content of single photons about an unknown polarisation rotation by a
  factor of over 200.\n\nReferences:\n[1] https://www.nature.com/articles/s
 41467-020-17559-w\n[2] https://arxiv.org/abs/1811.08046\n[3] https://arxiv
 .org/abs/2111.01194\n[4] https://iopscience.iop.org/article/10.1088/1751-8
 121/ac0289
LOCATION:MR2 Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR