BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Polariton graph simulators
DTSTART;VALUE=DATE-TIME:20190709T063000Z
DTEND;VALUE=DATE-TIME:20190709T070000Z
DTSTAMP;VALUE=DATE-TIME:20210122T232254Z
UID:indico-contribution-72816@indico.in2p3.fr
DESCRIPTION:Speakers: Natalia G. Berloff (Department of Applied Mathemati
cs and Theoretical Physics\, University of Cambridge and Skolkovo Institut
e of Science and Technology )\nRecently several gain-dissipative platforms
based on the networks of optical parametric oscillators\, coupled lasers
and various non-equilibrium Bose-Einstein condensates such as exciton-pola
riton\, atomic and photon condensates have been proposed as analogue Ham
iltonian simulators for solving large-scale hard optimisation problems. Su
ch platforms use an approach to finding the global minimum of spin Hamilto
nians which is different from quantum annealers or quantum computers. Howe
ver\, in general\, the parameters of such spin Hamiltonians when realised
by gain-dissipative simulators depend on the node occupancies that are no
t known a priori\, which limits the simulators applicability to the classe
s of problems easily solvable by classical computations. I show how to ov
ercome this difficulty and formulate algorithms for solving the NP-hard
large-scale optimisation problems such as constant modulus continuous qua
dratic optimisation and quadratic binary optimisation for any general matr
ix. To solve such problems any gain-dissipative simulator has to impleme
nt a feedback mechanism for the dynamical adjustment of the gain and coupl
ing strengths\, so that occupancy of each node is the same. I will illust
rate the work of such a simulator using the polariton graph platform that
we recently realised in experiments.\n\nhttps://indico.in2p3.fr/event/167
92/contributions/72816/
LOCATION:Cité des Congrès Salle GH
URL:https://indico.in2p3.fr/event/16792/contributions/72816/
END:VEVENT
END:VCALENDAR