BEGIN:VCALENDAR
PRODID:-//Brown Bear Software//Calcium 4.01//EN
VERSION:2.0
METHOD:PUBLISH
BEGIN:VEVENT
SUMMARY:Model-Based Derivative-Free Optimization with Unrelaxable Constraints
UID:x-1670-Calcium@vm-mp1-int
DTSTAMP:20220124T170833Z
DTEND:20210913T180000Z
CATEGORIES:PhD Thesis Presentation
ORGANIZER:MAILTO:Calcium@localhost.localdomain
DESCRIPTION:Abstract:\nWe develop model-based derivative-free trust-region algorithms for constrained optimization.\nOur algorithms are designed for problems in which infeasible points do not provide function values for either the objective or constraints. This restriction poses interesting challenges for ensuring the sample set is well-poised\, \nmeaning the relative positions of sample points ensure the interpolated model functions approximate the true functions well.\nTo address these challenges\, our algorithm constructs feasible ellipsoidal trust regions from which to choose sample points.\nThe algorithm is first developed for linear constraints and then extended to non-linear constraints.\n\nFor non-linear constraints\, there is no way to avoid some infeasible function evaluations.\nHowever\, the algorithm averts some such evaluation attempts by buffering the feasible region with second-order cones for each nearly-active constraint. Under reasonable assumptions\, this buffered region is feasible for sufficiently small trust region radii.\nThis ensures that the criticality measure for the iterates generated by our algorithm converges to zero.\nSpeaker: : Trever Hallock\nAffiliation: : \nLocation: : see email for zoom link
DTSTART:20210913T160000Z
END:VEVENT
END:VCALENDAR