Article,

An approach to guaranteeing generalisation in neural networks

, and .
Neural Networks, 14 (8): 1035 - 1048 (2001)
DOI: https://doi.org/10.1016/S0893-6080(01)00061-2

Abstract

A novel approach to generalisation is presented that is able, under certain circumstances, to guarantee the generalisation to binary-output data for which no targets have been given. The basis of the guarantee is the recognition of a persistent global minimum error solution. An empirical test for whether the guarantee holds is provided which uses a technique called target reversal. The technique employs two neural networks whose convergence using opposing targets signals validity of the guarantee.

Tags

Users

  • @garypolhill

Comments and Reviews