BEGIN:VCALENDAR
VERSION:2.0
PRODID:Icfo
X-PUBLISHED-TTL:P1W
BEGIN:VEVENT
UID:69ee030eb6be8
DTSTART:20221014T120000Z
SEQUENCE:0
TRANSP:OPAQUE
LOCATION:Auditorium
SUMMARY:ICFO | KORBINIAN KOTTMANN
CLASS:PUBLIC
DESCRIPTION:We perform quantum simulation on classical and quantum computer
 s and set up a machine learning framework in which we can map out phase di
 agrams of known and unknown quantum manybody systems in an unsupervised fa
 shion. The classical simulations are done with state-of-the-art tensor net
 work methods in one and two spatial dimensions. For one dimensional system
 s\, we utilize matrix product states (MPS) that have many practical advant
 ages and can be optimized using the efficient density matrix&nbsp\; enorma
 lization group (DMRG) algorithm. The data for two dimensional systems is o
 btained from entangled projected pair states (PEPS) optimized via imaginar
 y time evolution. Data in form of observables\, entanglement spectra\, or 
 parts of the State vectors from these simulations\, is then fed into a dee
 p learning (DL) pipeline where we perform anomaly detection to map out the
  phase diagram. We extend this notion to quantum computers and introduce q
 uantum variational anomaly detection. Here\, we ˝rst simulate the ground 
 state and then process it in a quantum machine learning (QML) manner. Both
  simulation and QML routines are performed on the same device\, which we d
 emonstrate both in realistic simulation and on a physical quantum computer
  hosted by IBM.\n&nbsp\;\nThesis Directors: Prof Dr. Antonio Ac&iacute\;n 
 and Prof. Dr. Maciej Lewenstein
DTSTAMP:20260426T122030Z
END:VEVENT
END:VCALENDAR