BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Neural network quantum states\, from lattice models to quantum che
 mistry and quantum computing - Guglielmo Mazzola\, IBM Research
DTSTART:20200928T153000Z
DTEND:20200928T163000Z
UID:TALK151405@talks.cam.ac.uk
CONTACT:Bingqing Cheng
DESCRIPTION:In this seminar I will present a fairly recent application of 
 machine learning to quantum physics\, more specifically\, the task to mach
 ine learn many-body quantum states.\n\nFirst I will introduce the seminal 
 work [1] where a variational representation of quantum states based on art
 ificial neural networks has been devised. \n\nThen I will show how such co
 mpact representation can be used to perform unsupervised learning type of 
 tasks\, such as quantum state tomography [2]\, i.e. how to characterise a 
 quantum state from a limited number of simple experimental measurements. A
 s an example\, I will show how this can benefit already existing quantum t
 echnologies\, i.e. in providing order-of-magnitude speed-up for certain ta
 sks ubiquitous in variational quantum computation [3].\nFinally I will bri
 efly overview very recent efforts to solve the Schrödinger equation for F
 ermionic systems with shallow and deep neural networks [4-5].\n\n\n[1] Car
 leo & Troyer\, Science 355 (6325)\, 602-606 (2017)\n\n[2] Torlai et. al. N
 ature Physics 14 (5)\, 447-450 (2018)\n\n[3] Torlai et. al. Physical Revie
 w Research 2 (2)\, 022060 (2020)\n\n[4] Choo et. al. Nature communications
  11 (1)\, 1-7 (2020)\n\n[5] Pfau et. al. arXiv:1909.02487 (2019)
LOCATION:virtual ZOOM meeting ID: 263 591 6003\, https://zoom.us/j/2635916
 003
END:VEVENT
END:VCALENDAR