BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//wp-events-plugin.com//7.2.3.1//EN
TZID:America/New_York
X-WR-TIMEZONE:America/New_York
BEGIN:VEVENT
UID:516@fds.yale.edu
DTSTART;TZID=America/New_York:20240306T113000
DTEND;TZID=America/New_York:20240306T130000
DTSTAMP:20250916T142130Z
URL:https://fds.yale.edu/events/fds-colloquium-ankur-moitra-mit-learning-f
 rom-dynamics/
SUMMARY:FDS Colloquium: Ankur Moitra (MIT) "Learning from Dynamics"
DESCRIPTION:Abstract:  Linear dynamical systems are the canonical model f
 or time series data. They have wide-ranging applications and there is a va
 st literature on learning their parameters from input-output sequences. Mo
 reover they have received renewed interest because of their connections to
  recurrent neural networks.\n\n\n\nBut there are wide gaps in our understa
 nding. Existing works have only asymptotic guarantees or else make restric
 tive assumptions\, e.g. that preclude having any long-range correlations. 
 In this work\, we give a new algorithm based on the method of moments that
  is computationally efficient and works under essentially minimal assumpti
 ons. Our work points to several missed connections\, whereby tools from th
 eoretical machine learning including tensor methods\, can be used in non-s
 tationary settings.\n\n\n\nBio: Ankur Moitra is a theoretical computer sc
 ientist\, and a major goal in his work is to give algorithms with provable
  guarantees for various problems in machine learning. He is a member of th
 e Theory of Computation group\, MachineLearning@MIT\, Foundations of Data 
 Science and the Center for Statistics.\n\n\n\nWebsite: https://people.csa
 il.mit.edu/moitra/\n
ATTACH;FMTTYPE=image/jpeg:https://fds.yale.edu/wp-content/uploads/2024/02/
 Moitra-Ankur-02b1fe5ae14c9162.jpeg
CATEGORIES:FDS Events,Statistics & Data Science
 Seminar,Colloquium,Seminar Series
END:VEVENT
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:STANDARD
DTSTART:20231105T010000
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
END:STANDARD
END:VTIMEZONE
END:VCALENDAR