We present an algorithm that induces a class of models with thin junction trees—models that are characterized by an upper bound on the size of.

We present an algorithm that induces a class of models with thin junction trees—models that are characterized by an upper bound on the size of the maximal ...

We say that a graphical model has a thin junction tree if its treewidth H is small. Our basic feature induction algorithm is a constrained variant of that ...

We present the first truly polynomial algorithm for learning the structure of bounded-treewidth junction trees -- an attractive subclass of probabilistic ...

In this paper, we propose a method for learning an attractive class of models: bounded treewidth junction trees. Those models permit both compact representation ...

In this technical report, I present a novel approximate filtering technique that avoids the quadratic space and time requirements of the Kalman filter technique ...

Using Bethe approximation of the likelihood, we transform the problem of finding a good junction tree separator into a minimum cut problem on a weighted graph.

Suzanna Becker is Associate Professor of Psychology and an Associate Member of the Department of Computing and Software at McMaster University.

Thin junction tree filtering (tjtf) is an approximate filtering technique for dynamic Bayesian networks. • tjtf is an assumed density filter where at each ...

We present the first truly polynomial algorithm for PAC-learning the structure of bounded-treewidth junction trees – an attractive subclass of probabilistic ...