Regular ArticleExponentiated Gradient versus Gradient Descent for Linear Predictors☆,☆☆
Under an Elsevier user license
open archive
Cited by (0)
- ☆
final manuscript received September 10, 1996
- ☆☆
An extended abstract appeared in “Proceedings of the 27 th Annual ACM Symposium on the Theory of Computing,” pp. 209–218, ACM Press, New York, May 1995.
- †
Supported by the Academy of Finland, Emil Aaltonen Foundation, and ESPRIT Project NeuroCOLT.
- ‡
Supported by NSF Grant IRI-9123692. E-mail: [email protected].
Copyright © 1997 Academic Press. All rights reserved.