This preview shows pages 1–10. Sign up to view the full content.
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: Outline of the Lecture Outline of the Lecture Energy Minimization Energy Minimization Energy Motion Aperture Problem Energy Motion Aperture Problem Regularization Regularization One Can Understand Some One Can Understand Some Network Computations as Network Computations as Energy Minimization Energy Minimization Swimming uses half centers, with cross inhibition ending mainly by local inhibition. The bifurcation diagram shows fixed points and limit cycles of varying frequencies as one modulates the tonic excitation to C neurons). David Marrs three levels of understanding computations. To understand the computation of a neural network, consider a simple feedforward case. A network like this is a resistive network, which, in steady state, minimizes electrical energy. Computations performed by neural networks can be expressed as energy minimization. The advantage to do so is emphasizing the computational strategy not the implementation. Example:...
View Full
Document
 Spring '09
 Grzywacz

Click to edit the document details