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

Neural Computation

August 1, 2001, Vol. 13, No. 8, Pages 1839-1861
(doi: 10.1162/08997660152469378)
© 2001 Massachusetts Institute of Technology
Recurrence Methods in the Analysis of Learning Processes
Article PDF (262.99 KB)
Abstract

The goal of most learning processes is to bring a machine into a set of “correct” states. In practice, however, it may be difficult to show that the process enters this target set. We present a condition that ensures that the process visits the target set infinitely often almost surely. This condition is easy to verify and is true for many well-known learning rules.To demonstrate the utility of this method, we apply it to four types of learning processes: the perceptron, learning rules governed by continuous energy functions, the Kohonen rule, and the committee machine.