Parameterization and Conditioning of Hinging Hyperplane Models
1996 (English)In: Proceedings of the 13th IFAC World Congress, 1996, Vol. 1, 227-232 p.Conference paper (Refereed)
Recently a new model class has emerged in the field of non-linear black-box modeling; the hinging hyperplane models. The hinging hyperplane model is closely related to the well known neural net models. In this contribution the parameterization of hinging hyperplane models is addressed. It is shown that the original setting is overparameterized and a new parameterization involving fewer parameters is suggested. Moreover, it is shown that there is nothing to loose in terms of negative effects in the numerical search when less parameters are used. The positive effects of a model class parameterized with less parameters is a decrease in computational complexity. In addition to the parameterization issues, another related question is discussed, namely if the estimation problem is ill-conditioned.
Place, publisher, year, edition, pages
1996. Vol. 1, 227-232 p.
Non-linear black-box modeling, System identification, Function approximation
IdentifiersURN: urn:nbn:se:liu:diva-93754OAI: oai:DiVA.org:liu-93754DiVA: diva2:628968
13th IFAC World Congress, San Francisco, CA, USA, June-July, 1996