BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Builderloo//Events//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
UID:event-721@builderloo.ca
DTSTAMP:20261026T145907Z
DTSTART:20261103T190000Z
DTEND:20261103T210000Z
SUMMARY:WICI Talk - Energy Efficient AI
DESCRIPTION:Details and registration: https://uwaterloo.ca/engineering/even
 ts/wici-talk-energy-efficient-ai\n\nOn Builderloo: https://builderloo.ca/e
 vents/721-wici-talk-energy-efficient-ai\n\nEnergy-Efficient AI: Why brains
  use much less power than AI\, and how this might help us improve current 
 AI Brains are incredibly energy-efficient\, with each connection using tho
 usands to millions of times less energy than a connection weight a modern 
 Large Language Model running on GPUs. This talk looks at why this is the c
 ase\, focusing on how brains differ from traditional digital hardware. Fur
 thermore\, we look at how modern understanding of how brains represent Tim
 e and Space (Grid Cells and Time Cells) has led to new AI algorithms that 
 improve efficiency. However\, to fully take advantage of these improvement
 s\, we also need to rethink how we design computer hardware\, leading to m
 ore "neuromorphic" approaches. About the Speaker Terry Stewart Associate R
 esearch Officer\, Centre for Theoretical Neuroscience\, University of Wate
 rloo Terry Stewart is currently an NRC Canada Associate Research Officer a
 nd adjunct professor with the Dept of Psychology at the University of Wate
 rloo. He is a former member of the Computational Neuroscience Research gro
 up at the University of Waterloo working with Chris Eliasmith on the devel
 opment of Spaun and the Nengo software.
LOCATION:RCH 209\, Roy Coutts Hall\, University of Waterloo\, 200 Universit
 y Ave W\, Waterloo\, ON
URL:https://uwaterloo.ca/engineering/events/wici-talk-energy-efficient-ai
END:VEVENT
END:VCALENDAR