BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
X-WR-CALNAME:Seminar @ Cornell Tech: Sarah Dean
X-WR-TIMEZONE:Eastern Time (US & Canada)
BEGIN:VEVENT
DTSTAMP:20260915T094116Z
UID:tag:localist.com\,2008:EventInstance_35910662965264
DTSTART:20210222T160000Z
DTEND:20210222T170000Z
DESCRIPTION:Via Zoom Video Conferencing\n\n \n\n"Reliable Machine Learning 
 in Feedback Systems"\n\nMachine learning techniques have been successful f
 or processing complex information\, and thus they have the potential to pl
 ay an important role in data-driven decision-making and control. However\,
  ensuring the reliability of these methods in feedback systems remains a c
 hallenge\, since classic statistical and algorithmic guarantees do not alw
 ays hold.\n\nIn this talk\, I will provide rigorous guarantees of safety a
 nd discovery in dynamical settings relevant to robotics and recommendation
  systems. I take a perspective based on reachability\, to specify which pa
 rts of the state space the system avoids (safety) or can be driven to (dis
 covery). For data-driven control\, we show finite-sample performance and s
 afety guarantees which highlight relevant properties of the system to be c
 ontrolled. For recommendation systems\, we introduce a novel metric of dis
 covery and show that it can be efficiently computed. In closing\, I discus
 s how the reachability perspective can be used to design social-digital sy
 stems with a variety of important values in mind.\n\n \n\nBIO:\n\nSarah is
  a PhD candidate in the Department of Electrical Engineering and Computer 
 Science at UC Berkeley\, advised by Ben Recht. She received her MS in EECS
  from Berkeley and BSE in Electrical Engineering and Math from the Univers
 ity of Pennsylvania. Sarah is interested in the interplay between optimiza
 tion\, machine learning\, and dynamics in real-world systems. Her research
  focuses on developing principled data-driven methods for control and deci
 sion-making\, inspired by applications in robotics\, recommendation system
 s\, and developmental economics. She is a co-founder of a transdisciplinar
 y student group\, Graduates for Engaged and Extended Scholarship in comput
 ing and Engineering\, and the recipient of a Berkeley Fellowship and a NSF
  Graduate Research Fellowship.
LOCATION:
SUMMARY:Seminar @ Cornell Tech: Sarah Dean
URL;VALUE=URI:https://events.cornell.edu/event/seminar_cornell_tech_sarah_d
 ean
CATEGORIES:Seminar
END:VEVENT
END:VCALENDAR