BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Penn Engineering Events - ECPv6.15.18//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Penn Engineering Events
X-ORIGINAL-URL:https://seasevents.nmsdev7.com
X-WR-CALDESC:Events for Penn Engineering Events
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20220313T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20221106T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20230312T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20231105T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20240310T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20241103T060000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20231024T110000
DTEND;TZID=America/New_York:20231024T120000
DTSTAMP:20260404T014803
CREATED:20230829T200102Z
LAST-MODIFIED:20230829T200102Z
UID:9537-1698145200-1698148800@seasevents.nmsdev7.com
SUMMARY:ESE Fall Seminar - "Acceleration by Stepsize Hedging"
DESCRIPTION:Can we accelerate convergence of gradient descent without changing the algorithm — just by optimizing stepsizes? Surprisingly\, we show that the answer is yes. Our proposed Silver Stepsize Schedule optimizes strongly convex functions in $k^{\log_p 2} = k^{0.7864}$ iterations\, where $p=1+\sqrt{2}$ is the silver ratio and $k$ is the condition number. This is intermediate between the textbook unaccelerated rate $k$ and the accelerated rate $\sqrt{k}$ due to Nesterov in 1983. The non-strongly convex setting is conceptually identical\, and we obtain an analogous accelerated rate $\eps^{-\log_p 2} = \eps^{-0.7864}$. We conjecture and provide partial evidence that these rates are optimal among all possible stepsize schedules. \nThe Silver Stepsize Schedule is an explicit non-monotonic fractal. Why should such stepsizes help? The core intuition is “hedging” between individually suboptimal strategies — short steps and long steps — since bad cases for the former are good cases for the latter\, and vice versa. Properly combining these stepsizes yields faster convergence due to the misalignment of worst-case functions. This talk is based on joint work with Pablo Parrilo that publishes and extends my 2018 Master’s Thesis — which established for the first time that judiciously chosen stepsizes can enable accelerated convex optimization. Prior to this thesis\, the only such result was for the special case of quadratics\, due to Young in 1953.
URL:https://seasevents.nmsdev7.com/event/ese-fall-seminar-title-tbd-5/
LOCATION:Glandt Forum\, Singh Center for Nanotechnology\, 3205 Walnut Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Colloquium
ORGANIZER;CN="Electrical and Systems Engineering":MAILTO:eseevents@seas.upenn.edu
END:VEVENT
END:VCALENDAR