| (3 intermediate revisions by 2 users not shown) | |||
| Line 1: | Line 1: | ||
| − | Published in ''Int. Journal for Numerical Methods in Engineering'' Vol. 75 (11), pp. 1341-1360, 2008<br /> | + | Published in ''Int. Journal for Numerical Methods in Engineering'' Vol. 75 (11), pp. 1341-1360, 2008<br /> |
doi: 10.1002/nme.2304 | doi: 10.1002/nme.2304 | ||
== Abstract == | == Abstract == | ||
In this work a conceptual theory of neural networks (NNs) from the perspective of functional analysis and variational calculus is presented. Within this formulation, the learning problem for the multilayer perceptron lies in terms of finding a function, which is an extremal for some functional. Therefore, a variational formulation for NNs provides a direct method for the solution of variational problems. | In this work a conceptual theory of neural networks (NNs) from the perspective of functional analysis and variational calculus is presented. Within this formulation, the learning problem for the multilayer perceptron lies in terms of finding a function, which is an extremal for some functional. Therefore, a variational formulation for NNs provides a direct method for the solution of variational problems. | ||
| − | This proposed method is then applied to distinct types of engineering problems. In particular a shape design, an optimal control and an inverse problem are considered. The selected examples can be solved analytically, which enables a fair comparison with the NN results | + | This proposed method is then applied to distinct types of engineering problems. In particular a shape design, an optimal control and an inverse problem are considered. The selected examples can be solved analytically, which enables a fair comparison with the NN results. |
<pdf>Media:Draft_Samper_903712874_7250_Lopez_et_al-2008-International_Journal_for_Numerical_Methods_in_Engineering.pdf</pdf> | <pdf>Media:Draft_Samper_903712874_7250_Lopez_et_al-2008-International_Journal_for_Numerical_Methods_in_Engineering.pdf</pdf> | ||
Latest revision as of 11:01, 12 February 2019
Published in Int. Journal for Numerical Methods in Engineering Vol. 75 (11), pp. 1341-1360, 2008
doi: 10.1002/nme.2304
Abstract
In this work a conceptual theory of neural networks (NNs) from the perspective of functional analysis and variational calculus is presented. Within this formulation, the learning problem for the multilayer perceptron lies in terms of finding a function, which is an extremal for some functional. Therefore, a variational formulation for NNs provides a direct method for the solution of variational problems. This proposed method is then applied to distinct types of engineering problems. In particular a shape design, an optimal control and an inverse problem are considered. The selected examples can be solved analytically, which enables a fair comparison with the NN results.
Share this document
Keywords
claim authorship
Are you one of the authors of this document?