BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:CQIF Seminar
SUMMARY:Quantum Generative Adversarial Networks for Learni
ng and Loading Random Distrubitions - Christa Zouf
al\, IBM Research/ETH
DTSTART;TZID=Europe/London:20191121T141500
DTEND;TZID=Europe/London:20191121T151500
UID:TALK131593AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/131593
DESCRIPTION:Quantum algorithms have the potential to outperfor
m their classical counterparts in a variety of tas
ks. The realization of the advantage often require
s the ability to load classical data efficiently i
nto quantum states. However\, the best known metho
ds require O(2^n) gates to load an exact represent
ation of a generic data structure into an n-qubit
state. This scaling can easily predominate the com
plexity of a quantum algorithm and\, thereby\, imp
air potential quantum advantage.\n\nOur work prese
nts a hybrid quantum-classical algorithm for effic
ient\, approximate quantum state loading. More pre
cisely\, we use quantum Generative Adversarial Net
works (qGANs) to facilitate efficient learning and
loading of generic probability distributions – im
plicitly given by data samples – into quantum stat
es. Through the interplay of a quantum channel\, s
uch as a variational quantum circuit\, and a class
ical neural network\, the qGAN can learn a represe
ntation of the probability distribution underlying
the data samples and load it into a quantum state
.\n\nThe loading requires O(poly (n)) gates and ca
n thus enable the use of potentially advantageous
quantum algorithms\, such as Quantum Amplitude Est
imation.\nWe implement the qGAN distribution learn
ing and loading method with Qiskit and test it usi
ng a quantum simulation as well as actual quantum
processors provided by the IBM Q Experience. Furth
ermore\, we employ quantum simulation to demonstra
te the use of the trained quantum channel in a qua
ntum finance application.
LOCATION:MR9\, Centre for Mathematical Sciences\, Wilberfo
rce Road\, Cambridge
CONTACT:Johannes Bausch
END:VEVENT
END:VCALENDAR