BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Urban Driving with Conditional Imitation Learning - Corina Gurau\,
  Researcher\, Wayve
DTSTART:20191121T130000Z
DTEND:20191121T140000Z
UID:TALK134026@talks.cam.ac.uk
CONTACT:Zohreh Shams
DESCRIPTION:*Abstract* \n\nEnd-to-end machine learning has enabled many br
 eakthroughs in computer vision over the last couple of years. At Wayve\, i
 t has enabled us to learn driving policies for autonomous vehicles without
  relying on HD-maps or hand-coded rules. In this talk I will present a sel
 f driving software stack for learning meaningful state representations wit
 h computer vision (semantics\, depth and motion) and driving policies usin
 g conditional imitation learning from expert drivers. These models can dri
 ve on public urban roads\, never seen before during training after learnin
 g from only 30 hours of real-world demonstrations.\n\n*Bio*\n\nCorina Gura
 u is a researcher at Wayve  applying machine learning and computer vision 
 to self-driving vehicles technology. She finished her DPhil at the Oxford 
 Robotics Institute\, where she worked on perception for robots operating o
 utdoors\, in diverse environmental conditions. Her thesis focused on under
 standing and remedying failure modes of ML-based vision systems. Before th
 at she finished a BSc in Computer Science at Jacobs University Bremen in G
 ermany.\n\n
LOCATION:Computer Laboratory\, William Gates Building\, Room FW26
END:VEVENT
END:VCALENDAR