M. Shimono et T. Yamakawa, Design and analysis of a nonequilibrium cross-coupled network with a detectable similarity measure, IEEE NEURAL, 11(1), 2000, pp. 57-68
In this paper, a nonequilibrium network which works as a dynamical associat
ive memory is designed. The design is based on a new similarity measure bet
ween any stored pattern and a state of the network. Although conventional s
imilarity measures, such as Hamming distance, direction cosine, and so on,
are not detectable in a cross-coupled network, the similarity measure propo
sed in this paper is. The new similarity measure is employed in our design,
The network should include the following properties in its output pattern
sequence, so that the dynamics of cross-coupled network mag be designed: 1)
Stored patterns are frequently associated in the dynamical association. 2)
The dynamical association is very robust against variation of distributed
parameters, Property 1) is achieved by introducing the next two operation m
odes with inverse N-shaped function into the dynamics of the proposed netwo
rk. 1) When the state of the network is close enough to a stored pattern at
a time step, the state is forced to evolve at the next time step, 2) The s
tate of the network converges to a stored one while it is not close to any
stored patterns, By considering these two operation modes, the frequency of
associating stored patterns is increased, The authors emphasize the proper
ty 2) which is very important for a silicon implementation of the proposed
network, In the silicon implementation, parameters of the network must be r
epresented by transistors, resistors, capacitors, and other electric compon
ents which exhibit variation in their characteristics. Thus the second prop
erty guarantees the easy silicon implementation of the nonequilibrium netwo
rk proposed in this paper.