Basic HTML version of Foils prepared 15 March 1996

Foil 20 Neural Net Compared to Elastic Net

From Physical Optimization and Physical Computation CPSP713 Case studies in Computational Science -- Spring Semester 1996. by Geoffrey C. Fox
Classical Neural Network Approach to TSP
    • hip= 1 : city p visited at time i
    • = 0 : otherwise
    • With Constraint that
    • for each i only one hip nonzero
Elastic net (roughly)
    • hi = p multistate neuron
    • (actually position in space of cities)
Simic showed how neural network (Hopfield-Tank) and elastic net came from making different mean field approximations to the same physical free energy

Northeast Parallel Architectures Center, Syracuse University, npac@npac.syr.edu

If you have any comments about this server, send e-mail to webmaster@npac.syr.edu.

Page produced by wwwfoil on Sun Feb 22 1998