BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Dynamical Networks for Agent Systems - Pedro Ortega\, Department o
 f Computer Science\, University of Chile
DTSTART:20060125T150000Z
DTEND:20060125T160000Z
UID:TALK4641@talks.cam.ac.uk
CONTACT:Phil Cowans
DESCRIPTION:The basic ideas for a novel class of recurrent neural networks
  that learn through active interaction with an unknown environment by read
 ing\ndata and reward information and by emitting actions through its I/O\n
 interface. The network's connections are fixed and no weights need to be a
 djusted. A neuron's activity is determined by one of two possible boolean 
 gates. Starting out with a uniform distribution over the two gates\, learn
 ing consists in assigning a gate to each neuron as evidence is gathered\, 
 thus "programming" the network on-the-fly.
LOCATION:Ryle Seminar Room\, Cavendish Laboratory
END:VEVENT
END:VCALENDAR