Title

Memory Capacity of a Neural Network

Authors

Takayuki Dan Kimura, Washington University in St. Louis

Document Type

Technical Report

Publication Date

1988-03-01

Filename

WUCS-88-11.pdf

DOI:

10.7936/K78050XQ

Technical Report Number

WUCS-88-11

Abstract

We propose to measure the memory capacity of a state machine by the numbers of discernible states, where two states are defined to be discernible if the machine manifests the identical input-output mapping in both states. According to the definition, a neuron network of n>0 inputs and one output, with an uncountable set of internal states, has the memory capacity of log2TF(n), where TF(n) is the number of different Boolean functions the network can realize with different synaptic weight and threshold values. It is shown that such a network with k>0 linear threshold units can realize at most 2k(n2+k2) Boolean functions and therefore the network has memory capacity of at most k(n2+k2) bits or 2k3 bits when n

Comments

Permanent URL: http://dx.doi.org/10.7936/K78050XQ

Recommended Citation

Kimura, Takayuki Dan, "Memory Capacity of a Neural Network" Report Number: WUCS-88-11 (1988). All Computer Science and Engineering Research.
https://openscholarship.wustl.edu/cse_research/768