Monthly
288 pp. per issue
6 x 9, illustrated
ISSN
0899-7667
E-ISSN
1530-888X
2014 Impact factor:
2.21

Neural Computation

Summer 1990, Vol. 2, No. 2, Pages 226-238.
(doi: 10.1162/neco.1990.2.2.226)
© 1990 Massachusetts Institute of Technology
Analysis of Neural Networks with Redundancy
Article PDF (501.81 KB)
Abstract

Biological systems have a large degree of redundancy, a fact that is usually thought to have little effect beyond providing reliable function despite the death of individual neurons. We have discovered, however, that redundancy can qualitatively change the computations carried out by a network. We prove that for both feedforward and feedback networks the simple duplication of nodes and connections results in more accurate, faster, and more stable computation.