Abstract
Previous studies of A* tree-searching have modeled heuristics as random variables. The average number of nodes expanded is expressed asymptotically in terms of distance to goal. The conclusion reached is that A* complexity is an exponential function of heuristic error: Polynomial error implies exponential complexity and logarithmic accuracy is required for polynomial complexity.
This paper eliminates simplifying assumptions of earlier studies. “Error” is replaced by a concept called “discrepancy”, a measure of the relative attractiveness to A* of a node for expansion when that node is compared with competing nodes on the solution path. According to our model, in order to have polynomial A* complexity, it is not necessary to have the “logarithmic accuracy” described in previous studies. Another way is for a heuristic's values to vary, or cluster, near a “central function” which grows at least as fast as distance to goal. Generally, logarithmic variation or less is adequate. For one class of heuristics considered, the faster this central function grows, the more is variation from it tolerated.
Similar content being viewed by others
References
A. Bagchi and A. Sen, Heuristic search in general tree structures: an average case analysis,Proc. 14th Annual ACM Computer Science Conf. (1986) pp. 309–318.
A. Bagchi and A. Sen, Average-case analysis of heuristic search in tree-like networks, in:Search in Artificial Intelligence, eds. L. Kanal and V. Kumar, (Springer, New York, 1988) pp. 131–165.
A. Bagchi and A. Mahanti, Search algorithms under different kinds of heuristics — a comparative study, J. ACM 30 (1983) 1–21.
A. Bagchi and A. Mahanti, Three approaches to heuristic search in networks, J. ACM 32 (1985) 1–27.
G. Brassard and P. Bratley,Algorithmics: Theory and Practice (Prentice-Hall, 1988).
H. Davis, Some effects of heuristic errors in A* tree-searching, Technical Report WSU-CS-89-01, Wright State University Department of Computer Science and Engineering (1989).
H. Davis, Cost-error relationships in A* tree-searching, J. ACM 37 (1990) 195–199.
H. Davis and S. Chenoweth, The mathematical modeling of heuristics, Ann. Math. Art. Int. 5 (1992).
J. Gaschnig, Performance measurement and analysis of certain search algorithms, Ph.D. Dissertation, Carnegie-Mellon University (1979).
P. Halmos,Measure Theory (Van Nostrand, New York, 1961).
N. Huyn, R. Dechter and J. Pearl, Probabilistic analysis of the complexity of A*, Art. Int. 15 (1980) 241–254.
J. Pearl,Heuristics: Intelligent Search Strategies for Computer Problem Solving (Addison-Wesley, 1984).
J. Pearl, Knowledge versus search: a quantitative analysis using A*, Art. Int. 20 (1983) 1–13.
I. Pohl, First results on the effect of error in heuristic search, in:Machine Intelligence 5, eds. B. Meltzer and D. Michie (American Elsevier, 1970) pp. 219–236.
Author information
Authors and Affiliations
Additional information
This research has been funded by NCR Corporation.
This research has been partially funded by a grant from NCR Corporation.
Rights and permissions
About this article
Cite this article
Chenoweth, S.V., Davis, H.W. New approaches for understanding the asymptotic complexity of A* tree searching. Ann Math Artif Intell 5, 133–162 (1992). https://doi.org/10.1007/BF01543474
Issue Date:
DOI: https://doi.org/10.1007/BF01543474