Artigo Revisado por pares

Using Neural Networks to Model Conditional Multivariate Densities

1996; The MIT Press; Volume: 8; Issue: 4 Linguagem: Inglês

10.1162/neco.1996.8.4.843

ISSN

1530-888X

Autores

Peter M. Williams,

Tópico(s)

Fault Detection and Control Systems

Resumo

Neural network outputs are interpreted as parameters of statistical distributions. This allows us to fit conditional distributions in which the parameters depend on the inputs to the network. We exploit this in modeling multivariate data, including the univariate case, in which there may be input-dependent (e.g., time-dependent) correlations between output components. This provides a novel way of modeling conditional correlation that extends existing techniques for determining input-dependent (local) error bars.

Referência(s)
Altmetric
PlumX