Title :
On the capacity of associative memories with linear threshold functions
Author_Institution :
Bell Lab., Murray Hill, NJ, USA
fDate :
7/1/1989 12:00:00 AM
Abstract :
Some important features of various known constructions of associative memories based on linear threshold functions are analyzed. Two important features are dealt with: (a) the ability to select an arbitrary set of desired memory vectors and design a network for this set; (b) the sizes and shapes of the domains of attraction of the desired memory vectors and their relation to various design parameters. The static capacity for randomly chosen desired memories is also analyzed. Two extremal examples of sets of desired memories are then analyzed in detail. For spectral schemes with randomly chosen O[N /ln N] memories, it is shown that almost all of the Hamming sphere around each memory is directly attracted
Keywords :
content-addressable storage; error correction; neural nets; Hamming sphere; associative memories; domains of attraction; error correction; linear threshold functions; memory vectors; neurons; randomly chosen desired memories; static capacity; Computer networks; Hopfield neural networks; IEL; Information systems; Linear programming; Neural networks; Neurons;
Journal_Title :
Information Theory, IEEE Transactions on
