BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Fundamental Barriers in Optimisation\, Statistics\, and Signal Pro
 cessing - Verner Vlacic (ETH)
DTSTART:20191212T150000Z
DTEND:20191212T160000Z
UID:TALK135154@talks.cam.ac.uk
CONTACT:Dr Hansen
DESCRIPTION:When solving optimisation problems such as linear programming 
 (LP)\, semidefinite programming (SDP)\, basis pursuit (BP)\, LASSO\, or tr
 aining neural networks\, we often discuss algorithms in terms of their abi
 lity to achieve an adequately small suboptimality of the objective functio
 n. In this talk we ask a different question: Do optimisation algorithms fi
 nd good approximations to exact optima? We will see that the answer to thi
 s question is surprisingly nuanced\, with implications for the theory of n
 umerical condition and complexity theory.
LOCATION:MR 14
END:VEVENT
END:VCALENDAR