Dear Rizwan,
the following papers might be of interest to you. They deal
with graphical models for nonlinear dependencies between
continuous variables, although they do not use sigmoid
dependency models.
"Discovering Structure in Continuous Variables using
Bayesian Networks"
Reimar Hofmann and Volker Tresp
NIPS*95
(http://wwwbrauer.informatik.tu-muenchen.de/~hofmannr/nips95_abstr.html)
"Nonlinear Markov Networks for Continuous Variables"
Reimar Hofmann and Volker Tresp
NIPS*97
(http://wwwbrauer.informatik.tu-muenchen.de/~hofmannr/nips97_abstr.html)
Best regards,
Reimar
