Apr. 19 2010 1:30PM

Apr. 19 2010 3:00PM

Compositional model theory represents an alternative approach to represent a multidimensional probability distribution. Every compositional model over a finite non-empty set of variables N is uniquely defined by its generating sequence - an ordered set of low-dimensional probability distributions. A generating sequence structure induces a system of conditional independence statements over N valid for every multidimensional distribution represented by a compositional model with the structure.
The equivalence problem is how to characterize whether all independence statements induced by structure P are induced by a second structure P' and vice versa. This problem can be solved in several ways. I will represent a partial solution of so called direct characterization of equivalence problem which is based on several structure properties necessary and sufficient for equivalence of the considered structures. Several properties invariant for equivalent structures will be discussed and introduced.

Pravidelný seminář Inteligentní systémy (Contact person: Tomáš Kroupa)