BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Rough Paths and more scalable data science - Terry Lyons (Oxford)
DTSTART:20220905T103000Z
DTEND:20220905T113000Z
UID:TALK178454@talks.cam.ac.uk
CONTACT:Jason Miller
DESCRIPTION:The basic principle of Rough Path Theory\, is that one can und
 erstand multi modal and oscillatory streams of data by considering their i
 mpact on non-linear controlled dynamical systems. \nThe approach leads to 
 the local description of these complex streams through signatures\; connec
 ting data science with tensor algebra. \nThese high order signatures of th
 e data are expensive to compute but can be re-used a huge number of times 
 in the training of a neural net without the need to re-compute the signatu
 re from the data.
LOCATION:MR9\, Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR