BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Data driven decision making and safety certificate synthesis - Kos
 tas Margellos (University of Oxford)
DTSTART:20251111T105000Z
DTEND:20251111T113000Z
UID:TALK238462@talks.cam.ac.uk
DESCRIPTION:\nData driven algorithms offer a natural framework to make dec
 isions in environments affected by uncertainty\, where uncertainty is repr
 esented by means of data. Neural networks constitute one class of such dat
 a driven decision making tools. However\, the ``learned&rsquo\;&rsquo\; de
 cisions are inherently random as they depend on the data used. In this tal
 k we discuss how tools from statistical learning theory based on the notio
 n of compression and randomized optimization offer a principled framework 
 to analyze the robustness properties of these learned decisions. Our resul
 ts build ``trust on data&rsquo\;&rsquo\;\, and accompany data driven solut
 ions with probabilistic robustness guarantees that capture their generaliz
 ation properties when it comes to new data\, not included in the learning/
 training process. We review recent advancements in this area that allow to
  boost performance based on a data outlier removal procedure. We then show
  how this methodology can be employed to build what will be referred to as
  safety informed neural networks\, that produce&nbsp\; safety and reachabi
 lity certificates for nonlinear dynamical systems\, accompanying them with
  prescribed probabilistic guarantees with respect to their validity.\n
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR