BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Alhussein Fawzi\, PhD: Proving inequalities with deep learning
DTSTART:20191104T160000
DTEND:20191104T170000
DTSTAMP:20260916T044002Z
UID:1aa2bbbee5aedd6b2ac6cee2d936b412afc4a0d9bf827c56f6d7068d
CATEGORIES:Conferences - Seminars
DESCRIPTION:Alhussein FAWZI is a research scientist at Google DeepMind in
  London\, working on making machine learning systems more robust and using
  machine learning for solving mathematical problems. He received his M.Sc.
  and PhD degrees from the Swiss Federal Institute of Technology (EPFL)\, S
 witzerland\, and spent one year as a postdoctoral scholar in the Computer 
 Science Department at UCLA. He received twice the IBM PhD fellowship. More
  information can be found in his website: http://www.alhusseinfawzi.info/
 \nI will consider in this talk the fundamental problem of searching for pr
 oofs of polynomial inequalities\, which has applications in control theory
 \, robotics\, geometry\, combinatorics\, and program verification to name 
 a few. While existing proof systems manipulating polynomial inequalities v
 ia elementary inference rules are known to be very powerful\, searching fo
 r proofs remains a major difficulty. I will introduce a deep reinforcement
  learning framework to search for a dynamic proof within these proof syste
 ms\, paying particular attention to incorporating inherent symmetries of t
 he problem as an inductive bias. By comparing our approach with powerful a
 nd widely-studied linear programming hierarchies based on static proof sys
 tems\, I will show that the proposed method reduces the size of the linear
  program by several orders of magnitude while also improving performance. 
 I will finally conclude with general remarks on the emerging field of usin
 g machine learning to solve mathematical problems.\n 
LOCATION:ELD 020 https://plan.epfl.ch/?room==ELD%20020
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR