Heavy-tailed distributions in combinatorial search
Document Type
Conference Proceeding
Publication Date
1-1-1997
Abstract
Combinatorial search methods often exhibit a large variability in performance. We study the cost profiles of combinatorial search procedures. Our study reveals some intriguing properties of such cost profiles. The distributions are often characterized by very long tails or “heavy tails”. We will show that these distributions are best characterized by a general class of distributions that have no moments (i.e., an infinite mean, variance, etc.). Such non-standard distributions have recently been observed in areas as diverse as economics, statistical physics, and geophysics. They are closely related to fractal phenomena, whose study was introduced by Mandelbrot. We believe this is the first finding of these distributions in a purely computational setting. We also show how random restarts can effectively eliminate heavy-tailed behavior, thereby dramatically improving the overall performance of a search procedure.
Identifier
84948994516 (Scopus)
ISBN
[3540637532, 9783540637530]
Publication Title
Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics
External Full Text Location
https://doi.org/10.1007/bfb0017434
e-ISSN
16113349
ISSN
03029743
First Page
121
Last Page
135
Volume
1330
Recommended Citation
Gomes, Carla P.; Selman, Bart; and Crato, Nuno, "Heavy-tailed distributions in combinatorial search" (1997). Faculty Publications. 16910.
https://digitalcommons.njit.edu/fac_pubs/16910
