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

Neural Computation

January 2009, Vol. 21, No. 1, Pages 1-8
(doi: 10.1162/neco.2009.09-07-615)
© 2008 Massachusetts Institute of Technology
Long-Range Out-of-Sample Properties of Autoregressive Neural Networks
Article PDF (66.19 KB)
Abstract

We consider already-trained discrete autoregressive neural networks in their most general representations, with the exclusion of time-varying input though, and we provide tight sufficient conditions and elementary proofs for the existence of an attractor, uniqueness, and global convergence. Those conditions can be used as easy-to-check criteria when convergence (or not) of long-range predictions is desirable.