When searching for prey, many predator species exhibit a remarkable behavior: after prey capture, the predators promptly engage in `area-restricted search', probing for consecutive captures nearby. Biologists have been surprised with the efficiency and adaptability of this search strategy to a great number of habitats and prey distributions. We propose to synthesize a similar search strategy for the massively multimodal problems of combinatorial optimization. The predatory search strategy restricts the search to a small area after each new improving solution is found. Subsequent improvements are often found during area-restricted search. Results of this approach to Gate Matrix Layout, an important problem arising in VLSI architectures, are presented. Compared to established methods over a set of benchmark circuits, predatory search is able to either match or outperform the best known layouts. Additional remarks address the relation of predatory search to the `big-valley' hypothesis and...

. We use the quasi-ordering of substructure relations such as induced and weak subgraph, induced suborder, graph minor or subformula of a CNF formula to obtain recognition algorithms for hereditary properties that are fast on average. The ingredients needed besides inheritance are independence of the occurrence of small substructures in a random input and the existence of algorithms for recognition that are at most exponential. 1 Introduction and Overview The aim of this paper is to show that a certain type of classes of recognition problems allows the design of simple and straight forward algorithms with low polynomial or even constant expected running time. This is done for the case that the property in question is hereditary w.r.t. some substructure relation ¯, i.e. if the set of objects is a quasi-order, qo, by means of ¯ and the objects with the property form a lower ideal in that qo. The only additional properties needed to show the existence of such algorithms that are fast in ...

This paper was written during a visit of the second author to LIP-IMAG/ENS Lyon. 2 Afonso G. Ferreira and Siang W. Song

We consider the development of practical algorithms based on the theory of graph minors. Although an exact decision algorithm using this approach would generally require testing to ensure the absence of a prohibitively large number of obstructions, approximate algorithms can be designed that test for only a few obstructions. Efficient self-reduction strategies can then be incorporated to approximate a solution to the problem at hand. In this paper, we investigate a prototypical problem, that of finding a three-track gate matrix layout for VLSI circuits. We design a streamlined test for a half dozen of the densest obstructions, thereby approximating an exact algorithm that would require over one hundred such tests, many of which appear very difficult. In this effort, we have also built a software package to automate the use of our tools. Experimental results obtained using this package demonstrate that it is extremely fast and suggest that, despite its theoretical limitations, its resul...

Abstract not found