Incremental Thin Junction Trees for Dynamic Bayesian networks
Frank Hutter · 2004
1 In this paper, we study the relationship between the thin junction tree filter (TJTF) [Pas03] and the Boyen-Koller (BK) algorithm [BK98a] for approximate inference in discrete dynamic Bayesian networks. First, we review the TJTF for discrete networks and cast the BK algorithm as a special case of TJTF. Then, we employ a TJTF to automat-ically compute conditionally independent clusters for the BK algo-rithm. Theoretical work by Boyen and Koller [BK99] showed that using conditionally independent clusters strongly improves BK’s error bounds, and we demonstrate that the theoretical results carry over to practice. We achieve a contract anytime algorithm which is superior to BK with marginally independent clusters and faster than TJTF in its general form. 1