Intelligent Search Techniques Research Articles

Do intelligent configuration search techniques outperform random search for large molecules?

AbstractWe compare three global configuration search methods on a scalable model problem to measure relative performance over a range of molecule sizes. Our model problem is a 2‐D polymer composed of atoms connected by rigid rods in which all pairs of atoms interact via Lennard–Jones potentials. The global minimum energy can be calculated analytically. The search methods are all hybrids combining a global sampling algorithm with a local refinement technique. The sampling methods are simulated annealing (SA), genetic algorithms (GA), and random search. Each of these uses a conjugate gradient (CG) routine to perform the local refinement. Both GA and SA perform progressively better relative to random search as the molecule size increases. We also test two other local refinement techniques in addition to CG, coupled to random search as the global method. These are simplex followed by CG and simplex followed by block‐truncated Newton. For small problems, the addition of the intermediate simplex step improved the performance of the overall hybrid method. © 1992 John Wiley & Sons, Inc.

A dynamic selection algorithm for globally optimal subsets

A new subset selection algorithm is proposed, based on artificial intelligence search techniques. It selects the globally best subset with the optimal size from an element set without exhaustive enumeration. Existing procedures either do not guarantee the optimality of the selected subset or do not determine optimal subset size. Exhaustive searches, on the other hand, are generally computationally unfeasible. The adoptability of the present algorithm is proved. The experiments, in the applications of the algorithm to clustering and feature selection, show its effectiveness.

Unidirectional and bidirectional search algorithms

Four classifications of artificial intelligence search techniques are discussed: unidirectional uniprocessor, bidirectional uniprocessor, unidirectional multiprocessor, and bidirectional multiprocessor search techniques. Wave-shaping PBA* (WS-PBA*) and search-space-clustering PBA*, (SSC-PBA*), two bidirectional AI search techniques, are compared. It is concluded that by maintaining a small number of formed clusters SSC-PBA* will be significantly faster than major wave-shaping bidirectional search algorithms. >

Future design environments for control engineering

The relationship between user needs and the potential of various computing techniques to satisfy them is examined for an interactive computing environment for work in the field of automatic control. The purpose is to stimulate the discussion of research directions in this field; practical guidelines or recipes for the user are not provided. The material presented is a mixture of surveys of existing work, analyses of needs and possibilities, and proposals and conclusions for further development. Particular stress is placed on the roles of language development, interface development, on conceptual models of the design process; the relative roles of man and machine; and expert systems, and the use of intelligent search techniques.

Future Design Environments for Control Engineering

The relationship between user needs and the potential of various computing techniques to satisfy them is examined for an interactive computing environment for work in the field of automatic control. The purpose is to stimulate the discussion of research directions in this field; practical guidelines or recipes for the user are not provided. The material presented is a mixture of surveys of existing work, analyses of needs and possibilities, and proposals and conclusions for further development. Particular stress is placed on the roles of language development, and interface development; on conceptual models of the design process, the relative roles of man and machine, and expert systems; and on the use of intelligent search techniques.