BEGIN:VCALENDAR VERSION:2.0 PRODID:researchseminars.org CALSCALE:GREGORIAN X-WR-CALNAME:researchseminars.org BEGIN:VEVENT SUMMARY:Asen L. Dontchev (University of Michigan) DTSTART:20210419T133000Z DTEND:20210419T143000Z DTSTAMP:20260513T222316Z UID:OWOS/44 DESCRIPTION:Title: Se nsitivity Analysis without Derivatives\nby Asen L. Dontchev (Universit y of Michigan) as part of One World Optimization seminar\n\n\nAbstract\nTh e classical sensitivity analysis developed in the early days of optimizati on and control revolves around determining derivatives of optimal values a nd solutions with respect to parameters in the problem considered. In pro blems with constraints however\, (standard) differentiability typically fa ils. The idea to obtain implicit function theorems without differentiabil ity goes back to Hildebrandt and Graves in their paper from 1927 and has been developed for optimization problems in the 1980s. \nIn this talk some major developments in sensitivity analysis of optimization problems in the last several decades are outlined. Estimates for solution depende nce on various perturbations are derived based on regularity properties o f mappings involved in the description of the problem. Applications to ma thematical programming\, numerical optimization and optimal control illu strate the theoretical findings.\n\nThe address and password of the zoom r oom of the seminar are sent by e-mail on the mailinglist of the seminar on e day before each talk\n LOCATION:https://researchseminars.org/talk/OWOS/44/ END:VEVENT END:VCALENDAR