BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Training deep quantum neural networks - Kerstin Beer (Leibniz Univ
 ersität Hannover)
DTSTART:20211015T100000Z
DTEND:20211015T110000Z
UID:TALK163024@talks.cam.ac.uk
CONTACT:40340
DESCRIPTION:Machine learning\, particularly as applied to deep neural netw
 orks via the back-propagation algorithm\, has brought enormous technologic
 al and societal change. With the advent of quantum technology it is a cruc
 ial challenge to design quantum neural networks for fully quantum learning
  tasks. In my talk I will present a truly quantum analogue of classical ne
 urons and explain how to use it to form a quantum feed-forward neural netw
 orks capable of universal quantum computation. For training these networks
  we use the fidelity as a cost function and benchmark the proposal for the
  quantum task of learning an unknown unitary operation. We find remarkable
  generalization behavior and robustness to noisy training data.  My talk w
 ill be based on a recent work of us [1]. For digging deeper in to the topi
 c after the talk I would recommend reading about finding an optimal lower 
 bound on the probability that such a trained network gives an incorrect ou
 tput for a random input [2] and about considering graph-structured quantum
  data for training our quantum neural networks [3]. \n\n\n[1] https://www.
 nature.com/articles/s41467-020-14454-2\n[2] https://arxiv.org/abs/2003.141
 03\n[3] https://export.arxiv.org/abs/2103.10837\n\n\nWhere: Virtually on Z
 oom https://us02web.zoom.us/j/88908652048?pwd=MDV3N3k0YnNWMlhKOEk1NDZlUEta
 UT09
LOCATION:Virtually\, at Zoom
END:VEVENT
END:VCALENDAR