M. Bengtsson. Higher order artificial neural networks. In T. Kohonen, K. Makisara, O. Simula, and J. Kangas, editors, Artificial Neural Networks. Proceedings of ICANN-91, pages 169--174, Amsterdam, 1991. North-Holland. Espoo, Finland, June 24-28, 1991. Home/Search Document Not in Database Summary Related Articles |
This paper is cited in the following contexts:
A Bayesian Neural Network Model With Extensions - Holst, Lansner (1993) (1 citation) (Correct)
....than having two two order columns, and it consumes considerably more memory, this can be a serious problem in some applications. Thus in many situations it would be more convenient if the columns were allowed to overlap. An important special case where such overlap occurs is a higher order network [22, 1]. In e.g. a second order network all second order events are represented, which in terms of this model means that we have one column for each possible pair of primary features. Still higher orders are defined similarly. These higher order networks can be considered as compromises ranging from ....
M. Bengtsson. Higher order artificial neural networks. In T. Kohonen, K. Makisara, O. Simula, and J. Kangas, editors, Artificial Neural Networks. Proceedings of ICANN-91, pages 169--174, Amsterdam, 1991. North-Holland. Espoo, Finland, June 24-28, 1991.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC