BEGIN:VCALENDAR

VERSION:2.0

PRODID:-//wp-events-plugin.com//6.6.4.4//EN

TZID:Asia/Jerusalem

X-WR-TIMEZONE:Asia/Jerusalem
BEGIN:VEVENT

UID:0-280@aerospace.technion.ac.il

DTSTART;TZID=Asia/Jerusalem:20201118T163000

DTEND;TZID=Asia/Jerusalem:20201118T173000

DTSTAMP:20230525T070751Z

URL:https://aerospace.technion.ac.il/events/graph-based-model-reduction-me
 thods-for-multi-agent-systems/

SUMMARY:Graph-Based Model Reduction Methods  for Multi-Agent Systems
DESCRIPTION:Lecturer:Noam Leiter\n Faculty:Department of Aerospace Engineer
 ing\n Institute:Technion – Israel Institute of Technology\n Location:htt
 ps://technion.zoom.us/j/96771910881\n Zoom: \n Abstract: \n Details: \n Mo
 del-order reduction is an essential tool for large-scale systems introduce
 d by modern Technologies for which full order simulation and controller de
 sign and implementation may be numerically and computationally infeasible.
  Model order reduction is a well established field of research in control 
 and systems theory. Of particular interest in recent years is the study of
  model reduction for multi-agent systems which are characterized by their 
 increasingly large scales.\nMulti-agent systems have an underlying network
  structure represented by graphs. For large-scale graphs\, combinatorial o
 perations can be performed to obtain reduced graph sizes. In general\, the
 re is an interest to understand how certain graph reduction operations pre
 serve spectral and combinatorial properties of the graph. We explore in th
 is talk certain graph reductions that satisfy an interlacing property betw
 een spectral properties of graphs represented by matrices. We show how two
  types of graph contractions\, cycle invariant contractions and node-remov
 al equivalent contractions\, lead to spectral interlacing of the normalize
 d-Laplacian and Laplacian graph matrices\, and we provide efficient algori
 thms for performing these contractions.\nWe then leverage these results on
  graph reductions to study model reduction of large-scale networked dynami
 c systems.  We show how graph contractions can be used to obtain reduced-
 order models with guarantees on the performance of the reduced model compa
 red to the original model.  In this direction\, we derive an a priori H2 
 error reduction bound for graph-based interlacing reduced models. We then 
 demonstrate these results on the classical controlled consensus model.\nZo
 om Meeting
CATEGORIES:Seminars
LOCATION:https://technion.zoom.us/j/96771910881

END:VEVENT

BEGIN:VTIMEZONE

TZID:Asia/Jerusalem

X-LIC-LOCATION:Asia/Jerusalem

BEGIN:STANDARD

DTSTART:20201025T010000

TZOFFSETFROM:+0300

TZOFFSETTO:+0200

TZNAME:IST

END:STANDARD

END:VTIMEZONE
END:VCALENDAR