[comp.robotics] CMAC vs. other neural architectures

unhd (Roger Gonzalez ) (06/13/90)

I'm just starting to learn about CMAC memory (reading papers by Albus
and Miller) and would like any references or information pertaining to
comparisons between CMAC and standard nets (i.e.  BP), especially with
regards to robotic control.  I am much more familiar with the McClelland
and Rumelhart stuff than CMAC; is it more closely related to Pentti
Kanerva's sparse distributed memory? I may be able to work with a
experimental CMAC memory board, if I can write up a good proposal. 

Thanks,
Roger

-- 
UUCP:   ..!uunet!unhd!rg      | USPS: Marine Systems Engineering Laboratory
BITNET: r_gonzalez at unhh    |       University of New Hampshire
PHONE:  (603) 862-4600        |       Marine Programs Building
FAX:    (603) 862-4399        |       Durham, NH  03824-3525