Backpropagation separates when perceptrons do

Research output: Contribution to conferencePaperpeer-review

21 Scopus citations


Consideration is given to the behavior of the least-squares problem that arises when one attempts to train a feedforward net with no hidden neurons. It is assumed that the net has monotonic nonlinear output units. Under the assumption that a training set is separable, that is, that there is a set of achievable outputs for which the error is zero, the authors show that there are no nonglobal minima. More precisely, they assume that the error is of a threshold least-mean square (LMS) type, in that the error function is zero for values beyond the target value. The authors' proof gives, in addition, the following stronger result: the continuous gradient adjustment procedure is such that from any initial weight configuration a separating set of weights is obtained in finite time. Thus they have a precise analog of the perceptron learning theorem. The authors contrast their results with the more classical pattern recognition problem of threshold LMS with linear output units.

Original languageEnglish (US)
Number of pages4
StatePublished - 1989
EventIJCNN International Joint Conference on Neural Networks - Washington, DC, USA
Duration: Jun 18 1989Jun 22 1989


OtherIJCNN International Joint Conference on Neural Networks
CityWashington, DC, USA

All Science Journal Classification (ASJC) codes

  • Engineering(all)


Dive into the research topics of 'Backpropagation separates when perceptrons do'. Together they form a unique fingerprint.

Cite this