%0 Journal Article
%T Discovering Multiple Constraints that are Frequently Approximately Satisfied
%A Geoffrey E. Hinton
%A Yee Whye Teh
%J Computer Science
%D 2013
%I arXiv
%X Some high-dimensional data.sets can be modelled by assuming that there are many different linear constraints, each of which is Frequently Approximately Satisfied (FAS) by the data. The probability of a data vector under the model is then proportional to the product of the probabilities of its constraint violations. We describe three methods of learning products of constraints using a heavy-tailed probability distribution for the violations.
%U http://arxiv.org/abs/1301.2278v1