© The Institution of Electrical Engineers
A correlation significance technique is introduced, which uses the significances between the training associations and between the neurons of the different fields, such that the storage capacity of Kosko bidirectional associative memories (BAMs) is improved. Based on the gradient descent method, an error function is constructed such that the maximal number of stored associations which are successfully recalled can be obtained. Finally computer simulations demonstrate that the proposed method has much better storage capacity than the existing methods.
References
-
-
1)
-
M.J. Sohn ,
S.Y. Shin
.
Reduction of interconnections in quadratic associative memory using pair significance.
Electron. Lett.
,
17 ,
1669 -
1670
-
2)
-
Hopfield, J.J.: `Neurons with graded response have collective computational properties like those of two-state neurons', Proc. Nat. Acad. Sci., 1984, U.S., 81, p. 3088–3092.
-
3)
-
Y.F. Wang ,
J.B. Cruz ,
J.H. Mulligan
.
Guaranteed recall of all training pairs for bidirectional associative memory.
IEEE Trans.
,
559 -
567
-
4)
-
B. Kosko
.
Bidirectional associative memories.
IEEE Trans.
,
49 -
60
-
5)
-
Haines, K., Hecht-Nielsen, R.: `A BAM with increased information storage capacity', Proc. IJCNN 88, 1988, 1, p. 181–190.
-
6)
-
S.-Y. Lee ,
J.-S. Jang ,
S.-Y. Shin ,
C.S. Shim
.
Optical implementation of associative memory with controlled bit-significance.
Appl. Opt.
,
1921 -
1923
-
7)
-
Y.F. Wang ,
J.B. Cruz ,
J.H. Mulligan
.
Two coding strategies for bidirectional associative memory.
IEEE Trans.
,
81 -
92
http://iet.metastore.ingenta.com/content/journals/10.1049/el_19930461
Related content
content/journals/10.1049/el_19930461
pub_keyword,iet_inspecKeyword,pub_concept
6
6