BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Learning to navigate without a map (but with instructions) - Piotr
  Mirowski\, DeepMind
DTSTART:20190208T120000Z
DTEND:20190208T130000Z
UID:TALK115090@talks.cam.ac.uk
CONTACT:Andrew Caines
DESCRIPTION:Navigation is an important cognitive task that enables humans 
 and animals to traverse\, with or without maps\, over long distances in th
 e complex world. Such long-range navigation can simultaneously support sel
 f-localisation (“I am here”) and a representation of the goal (“I am
  going there”). For this reason\, studying navigation is fundamental to 
 the study and development of artificial intelligence\, and trying to repli
 cate navigation in artificial agents can also help neuroscientists underst
 and its biological underpinnings. This talk will cover our own journey to 
 understand navigation by building deep reinforcement learning agents\, sta
 rting from learning to control a simple agent that can explore and memoris
 e large 3D mazes. I will show how these artificial agents relate to naviga
 tion in the real world\, both through the study of the emergence of grid c
 ell representations in neural networks and by demonstrating that these age
 nts can navigate in Street View-based real world photographic environments
 . I will finally present a new task and preliminary results on navigating 
 in Street View by following language instructions.
LOCATION:FW26\, Computer Laboratory
END:VEVENT
END:VCALENDAR