BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Date iCal//NONSGML kigkonsult.se iCalcreator 2.20.2//
METHOD:PUBLISH
X-WR-CALNAME;VALUE=TEXT:Eventi DIAG
BEGIN:VTIMEZONE
TZID:Europe/Paris
BEGIN:STANDARD
DTSTART:20231029T030000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20240331T020000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:calendar.27559.field_data.0@www.diag.uniroma1.it
DTSTAMP:20260404T144436Z
CREATED:20231204T124737Z
DESCRIPTION:AbstractDirect Multisearch (DMS) is a class of methods for mult
 iobjective derivative-free optimization that has a well-established conver
 gence analysis and competitive computational implementations\, being often
  used as benchmark for new algorithms or in practical applications. From a
  theoretical point of view\, DMS was developed for continuous optimization
  with general constraints\, using an extreme barrier approach where only f
 easible points are evaluated. In this work\, we propose the integration of
  an inexact restoration filter approach in DMS\, to address optimization p
 roblems with general constraints. Like in any filter approach\, violations
  of the relaxable constraints are addressed as an additional objective tha
 t needs to be minimized. The inexact restoration approach attempts to reco
 ver feasibility when the poll center is infeasible. Under mild assumptions
 \, we prove that the so-called DMS-FILTER-IR algorithm generates feasible 
 and/or infeasible subsequences of iterates that converge to either a Paret
 o stationary point\, in the feasible case\, or to a Pareto stationary poin
 t for the problem that only considers the unrelaxable constraints\, potent
 ially serving as a Pareto stationary point of the original problem\, in th
 e infeasible case. We will detail the proposed algorithm\, provide theoret
 ical results on convergence\, and report numerical experiments that state 
 the good performance of this approach to address multiobjective problems w
 ith general constraints.
DTSTART;TZID=Europe/Paris:20231213T110000
DTEND;TZID=Europe/Paris:20231213T110000
LAST-MODIFIED:20231204T125456Z
LOCATION:Aula A3\, via Ariosto 25
SUMMARY:Prof. Everton J. Silva: An Inexact Restoration Direct Multisearch F
 ilter Approach to Constrained Optimization - Prof. Everton J. Silva
URL;TYPE=URI:https://www.diag.uniroma1.it/node/27559
END:VEVENT
END:VCALENDAR
