We present a learning-parameter adjustment algorithm, valid for a large class of learning rules in neural-network literature. The algorithm follows directly from a consideration of the statistics of the weights in the network. The characteristic behavior of the algorithm is calculated, both in a fixed and a changing environment. A simple example, Widrow-Hoff learning for statistical classification, serves as an illustration.

• Received 30 December 1991

DOI:https://doi.org/10.1103/PhysRevA.45.8885