BEGIN:VCALENDAR
VERSION:2.0
PRODID:researchseminars.org
CALSCALE:GREGORIAN
X-WR-CALNAME:researchseminars.org
BEGIN:VEVENT
SUMMARY:Armenak Petrosyan (Georgia institute of Technology)
DTSTART:20220507T140000Z
DTEND:20220507T150000Z
DTSTAMP:20260423T035821Z
UID:YMC/45
DESCRIPTION:Title: <a href="https://researchseminars.org/talk/YMC/45/">Har
 monic Analysis and Approximation Theory techniques in Machine Learning\, p
 art 2</a>\nby Armenak Petrosyan (Georgia institute of Technology) as part 
 of Yerevan Mathematical Colloquium\n\n\nAbstract\nIn this talk\, I will pi
 ck up my presentation from where I left off last time. The main challenge 
 I will address in this talk is to find small size shallow neural networks 
 that can be trained algorithmically and which achieve guaranteed approxima
 tion speed and precision. To maintain the small size we apply penalties on
  the weights of the network. We show that under minimal requirements\, all
  local minima of the resulting problem are well behaved and possess a desi
 rable small size without sacrificing precision.\n\nAdditionally\, I will p
 resent an overview of the topics in the upcoming Focus Program on Data Sci
 ence\, Approximation Theory\, and Harmonic Analysis at Fields Institute in
  Toronto\, Canada where I am one of the coorganizers. This will be held in
  a hybrid format from May 9 to June 10\, 2022. More details at http://www.
 fields.utoronto.ca/activities/21-22/data\n\nTalk host: Michael Poghosyan (
 YSU)\n
LOCATION:https://researchseminars.org/talk/YMC/45/
END:VEVENT
END:VCALENDAR