BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//wp-events-plugin.com//7.2.3.1//EN
BEGIN:VEVENT
UID:468@lincs.fr
DTSTART;TZID=Europe/Paris:20190710T150000
DTEND;TZID=Europe/Paris:20190710T160000
DTSTAMP:20190715T110052Z
URL:https://www.lincs.fr/events/talk-by-shirin-jalali/
SUMMARY:Towards theoretically-founded structure learning
DESCRIPTION:\n Solving most inference tasks\, such as denoising and
 linear&nbsp\;regression\, relies on exploiting the structure of the desired
 class of signals (e.g. images). &nbsp\;Traditionally\, such structures are
 discovered by&nbsp\;domain experts after extensive studies. &nbsp\;This has
 to a great extent&nbsp\;limited the discovery and application of complex
 structures that exist&nbsp\;in many signals of interest. Learning-based
 methods that automatically&nbsp\;recover the structure of the source from
 available training datasets&nbsp\;provide a promising alternative solution.
 &nbsp\;However\, for&nbsp\;continuous-valued signals\, learning the source
 distribution is extremely&nbsp\;challenging\, and theoretically-founded
 computationally-feasible&nbsp\;approaches are yet to be found. &nbsp\;In
 this talk\, I will discuss&nbsp\;recently-proposed structure learning
 methods that\, instead of learning&nbsp\;the full distribution of the
 source\, learn its key features that are&nbsp\;relevant to solving
 inference problems. &nbsp\;As I will discuss\, this method\,&nbsp\;while
 substantially reducing the computational complexity of the&nbsp\;structure
 learning task\, leads to asymptotically optimal
 learning-based&nbsp\;estimators.\n
CATEGORIES:Seminars
LOCATION:Paris-Rennes Room (EIT Digital)\, 23 avenue d'Italie\, 75013
 Paris\, France
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=23 avenue d'Italie\, 75013
 Paris\, France;X-APPLE-RADIUS=100;X-TITLE=Paris-Rennes Room (EIT
 Digital):geo:0,0
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/Paris
X-LIC-LOCATION:Europe/Paris
BEGIN:DAYLIGHT
DTSTART:20190331T030000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
END:VCALENDAR