Stochastic neural network training algorithms in the problem of determining transport coefficients of the medium (Q1281243)
From MaRDI portal
 | This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: Stochastic neural network training algorithms in the problem of determining transport coefficients of the medium |
scientific article; zbMATH DE number 1266903
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Stochastic neural network training algorithms in the problem of determining transport coefficients of the medium |
scientific article; zbMATH DE number 1266903 |
Statements
Stochastic neural network training algorithms in the problem of determining transport coefficients of the medium (English)
0 references
23 March 1999
0 references
A method of reconstruction of diffusion and convection coefficients by means of measurements of electronic temperature in plasma is proposed. The method bases on the theory of laminated neural networks -- perceptrones. The advantage of this approach consists in its effective realisation by neural network itself without using external calculator. A stochastic algorithm for training the perceptrones based on the Monte-Carlo method is suggested. A comparative analysis of the method proposed and an iterative regularization method is presented.
0 references
