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

Neural Computation

November 15, 1997, Vol. 9, No. 8, Pages 1661-1665
(doi: 10.1162/neco.1997.9.8.1661)
© 1997 Massachusetts Institute of Technology
A Local Learning Rule That Enables Information Maximization for Arbitrary Input Distributions
Article PDF (44.43 KB)
Abstract

This note presents a local learning rule that enables a network to maximize the mutual information between input and output vectors. The network's output units may be nonlinear, and the distribution of input vectors is arbitrary. The local algorithm also serves to compute the inverse C−1 of an arbitrary square connection weight matrix.