BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Superpower Daily//AI Events//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Superpower Daily AI events
BEGIN:VEVENT
UID:c259b8db-69e8-4aa7-90f6-f51f9da753d0@superpowerdaily.com
DTSTAMP:20261107T140356Z
LAST-MODIFIED:20261107T140356Z
DTSTART:20261112T200000Z
DTEND:20261112T204500Z
SUMMARY:Beyond worst-case algorithms for learning
DESCRIPTION:The talk will explore how natural structure in data and learnin
 g tasks can enable efficient algorithms for problems believed to be comput
 ationally hard\, including linear classification\, sparse regression\, and
  graphical-model learning. It will highlight key results and broader direc
 tions for understanding when structure can overcome computational barriers
 .\nDetails: https://superpowerdaily.com/events/beyond-worst-case-algorithm
 s-for-learning-c259b8db-69e8-4aa7-90f6-f51f9da753d0\nRegister: https://kem
 pnerinstitute.harvard.edu/events/research-fellow-candidate-presentation-ta
 lk-title-tbd-8
LOCATION:Allston\, MA
URL:https://superpowerdaily.com/events/beyond-worst-case-algorithms-for-lea
 rning-c259b8db-69e8-4aa7-90f6-f51f9da753d0
STATUS:TENTATIVE
TRANSP:TRANSPARENT
END:VEVENT
END:VCALENDAR