We investigate the application of artificial neural networks (ANNs) for the
interpretation of Stokes profiles. We have employed ANNs to approximate th
e nonlinear inverse mapping between the Stokes profiles and some of the und
erlying atmospheric parameters. This approximate model is used in the follo
wing to carry out a fast non-iterative inversion of synthetic Stokes profil
es. We have used synthetic Stokes profiles of the photospheric infrared lin
e Fe I lambda 15648 to demonstrate that the ANNs are capable to yield accur
ate single valued estimates of the complete magnetic field vector, line-of-
sight (LOS) velocity, microturbulence, macroturbulence and the filling fact
or with exceptional speed. For a stratified atmosphere we also demonstrate
that these single valued parameters do represent very good averaged values
of the input stratification. To retrieve some of the temperature informatio
n encoded in the Stokes profiles we modeled a neural network classifier on
the basis of several semi-empirical model atmospheres (i.e. temperature and
pressure stratification). With this classifier we are able to determine th
e probability that a given Stokes profile has its origin from a particular
temperature stratification of a semi-empirical model.