Abstract

In this paper we propose a family of algorithms combining tree-clustering with conditioning that trade space for time. Such algorithms are useful for reasoning in probabilistic and deterministic networks as well as for accomplishing optimization tasks. By analyzing the problem structure it will be possible to select from a spectrum the algorithm that best meets a given time-space specification. 1 INTRODUCTION Topology-based algorithms for constraint satisfaction and probabilistic reasoning fall into two distinct classes. One class is centered on tree-clustering, the other on cycle-cutset decomposition. Tree-clustering involves transforming the original problem into a treelike problem that can then be solved by a specialized tree-solving algorithm [ Mackworth and Freuder, 1985; Pearl, 1986 ] . The tree-clustering algorithm is time and space exponential in the induced width (also called tree width) of the problem's graph. The transforming algorithm identifies subproblems that together ...

Citations

5662	Probabilistic reasoning in intelligent systems: networks of plausible inference - Pearl - 1988
1080	Local computations with probabilities on graphical structures and their application to expert systems (with discussion - Lauritzen, Spiegelhalter - 1988
837	Temporal constraint networks - Dechter, Meiri, et al. - 1991
360	Network-based heuristics for constraint satisfaction problems - Dechter, Pearl - 1987
316	Gaussian influence diagrams - Shachter, Kenley - 1989
298	Fusion, propagation, and structuring in belief networks - Pearl - 1986
293	Influence diagrams - Howard, Matheson - 2005
291	A neutral netlist of 10 combinational benchmark circuits and a target translator - Brglez, Fujiwara - 1985
267	Complexity of finding embeddings in a k-tree - Arnborg, Corneil, et al. - 1987
250	Enhancement schemes for constraint processing: Backjumping, learning and cutset decomposition - Dechter - 1990
247	The complexity of some polynomial network consistency algorithms for constraint satisfaction problems - Mackworth, Freuder - 1985
240	Bayesian updating in causal probabilistic networks by local computations - Jensen, Lauritzen, et al. - 1990
238	Bucket elimination: A unifying framework for probabilistic inference - Dechter - 1998
238	A Sufficient condition for backtrack-free search - Freuder
230	Tree clustering for constraint networks - Dechter, Pearl - 1989
229	Bucket elimination: a unifying framework for reasoning - Dechter - 1999
183	Efficient algorithms for combinatorial problems on graphs with bounded decomposability - a survey - Arnborg - 1985
153	Nonserial Dynamic Programming - Bertele, Brioschi - 1972
137	A sufficient condition for backtrack-bounded search - Freuder - 1985
130	Linear time algorithms for NP-hard problems restricted to partial k-trees, Discrete Applied Math - Proskurowski, Arnborg - 1989
101	Graphical models for preference and utility - Bacchus, Grove - 1995
95	Probability propagation - Shafer, Shenoy - 1990
53	Dynamic Variable Ordering in CSPs - Bacchus, Run - 1995
53	A su cient condition for backtrack-free search - Freuder - 1982
51	Model-based diagnosis using structured system descriptions - Darwiche
47	Fusion and propagation with multiple observations in belief networks - Peot, Shachter - 1991
46	A practical algorithm for finding optimal triangulations, in: AAAI’97: Proceedings of the 14th national conference on Artificial intelligence - Shoikhet, Geiger - 1997
44	A neutral netlist of 10 combinatorial benchmark circuits - Brglez, Fujiwara - 1985
42	Global conditioning for probabilistic inference in belief networks - Shachter, Andersen
34	Complexity of nding embeddings in a k-tree - Arnborg, Corneil, et al. - 1987
26	Local conditioning in Bayesian networks - D'iez - 1996
24	Diagnosing tree-decomposable circuits - Fattah, Dechter - 1995
24	Evaluating In uence Diagrams - Shachter - 1986
22	A probabilistic approach to hierarchical model-based diagnosis - Srinivas - 1994
21	An improved constraint propagation algorithm for diagnosis - ner, Pearl - 1987
15	Conditional algorithms for exact and approximate inference in causal networks - Darwiche - 1995
15	Some improvements to the Shenoy–Shafer and Hugin architecture for computing marginals - Schmidt, Shenoy - 1998
13	An evaluation of structural parameters for probabilistic reasoning: results on benchmark circuits - El-Fattah, Dechter - 1996
8	To guess or to think? hybrid algorithms for SAT - Rish, Dechter - 1996
5	A su#ciently fast algorithm for finding close to optimal junction trees - Becker, Geiger - 1996
5	A combination of cutset conditioning with clique-tree propagation in the pathfinder system - Suermondt, Cooper, et al. - 1990
3	Nested junction trees - Kjaerul - 1997
3	Fusion and proagation with multiple observations in belief networks - Peot, Shachter - 1992
3	Resolution vs. sat: two approaches to sat - Rish, Dechter - 2000
3	Probabilistic expert systems. Society for indutrial and Applied Mathematics - Shafer - 1996
2	Cyclic clustering: a compromise between tree-clustering and the cycle-cutset method for improving search e ciency - Jegou - 1990
2	Foundations of measurements, academic press - Krantz, Luce, et al. - 1976
1	Triangulation of graph-based algorithms giving small total state space - Kjaerul
1	Optimal junction trees - Jennsen, Jennsen - 1994
1	Enhancement schems for constraint processing incorporating, backjumping, learning and cutset decomposition. Arti cial Intelligence - Dechter - 1990