BEGIN:VCALENDAR
VERSION:2.0
METHOD:PUBLISH
CALSCALE:GREGORIAN
PRODID:iCalendar-Ruby
BEGIN:VEVENT
LOCATION:Tolman 5101
SEQUENCE:0
DTEND:20091113T130000
DTSTART:20091113T110000
UID:philosophy.berkeley.edu:events:562
DTSTAMP:20091124T185753
DESCRIPTION:One of the great unsolved questions in our field is how the hum
 an brain\, and simulations thereof\, can achieve the kind of common-sense u
 nderstanding that is widely believed to be essential for robust intelligenc
 e.  Many have argued that embodiment is important for developing common-sen
 se understanding\, but exactly how this occurs at a mechanistic level remai
 ns unclear.  In the process of building an embodied cognitive agent that le
 arns from experience in a virtual environment\, my colleagues and I have de
 veloped several insights into \nthis process.   First\, embodiment provides
  access to a rich\, continuous source of training signals that\, in conjunc
 tion with the proper neural structures\, naturally support the learning of 
 complex sensory-motor abilities.  Second\, there is an intriguing developme
 ntal cascade of learning\, where initial learning to fixate (foveate) a tar
 get enables subsequent learning to reach for that target\, and also to reco
 gnize it within a cluttered visual environment. Finally\, there \nare impor
 tant functional differences in the learning mechanisms required for differe
 nt brain areas and associated domains\, which converge well with bottom-up 
 biological data\, including that on spike- timing dependent plasticity (STD
 P).
SUMMARY:Institute of Cognitive and Brain Sciences\nRrandy O'Reilly\nFrom Sp
 ikes to Object Recognition and Beyond: Building an Embodied Brain
END:VEVENT
END:VCALENDAR