This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: A recurrent network is a feedforward network with a recurrent synaptic weight matrix. Some neuronal tissues are so massive and complex that network analysis is not too useful. Perception is a constructive process that depends on both the stimulus information and the mental structure of the perceiver. Even when network analysis is not useful, we can still comprehend aspects of brain functions with models that are more generic. David Marr’s three levels of understanding computations. In a blackbox model, we try to describe a system well enough to predict its responses without knowing what is inside the system. If the black box is linear, then we can describe the system fully with the impulse response, as any stimulus is a sum of impulses. The impulse response D(t) is the reaction to a very short stimulus at time zero. One can use this model to estimate (rest(t)) the response of a linear system to stimulus s(t): r est t ( ) = r + d τ D τ ( ) s t − τ ( ) ∞ ∫ In Temporal Difference learning, we use a discrete time variable t...
View
Full
Document
This note was uploaded on 06/08/2009 for the course BME 575L taught by Professor Grzywacz during the Spring '09 term at USC.
 Spring '09
 Grzywacz

Click to edit the document details