The RNA secondary structure prediction problem (2 RNA) is a critical one in molecular biology. Secondary structure can be determined directly by x-ray diffraction, but this is difficult, slow, and expensive. Moreover, it is currently impossible to crystallize most RNAs. Mathematical models for prediction have therefore been developed and these have led to serial (and some parallel) computer algorithms, but these too are expensive in terms of computation time. The general solution has asymptotic running time exponential in N (i.e., proportional to 2 N), where N is the length of the RNA sequence. Serial approximation algorithms which employ heuristics and make strong assumptions are significantly faster, on the order of N 3 or N 4, but their predictive success rates are low -- often less than 40 percent -- and even these algorithms can run for days when processing very long (thousands of bases) RNA sequences. Neural network algorithms that perform a multiple constraint satisfaction search using a massively parallel network of simple processors may provide accurate and very fast solutions.

Shortly after Watson and Crick's discovery of the structure of DNA, and at about the same time that the genetic code and the essential facts of gene expression were being elucidated, the field of linguistics was being similarly revolutionized by the work of Noam Chomsky [Chomsky, 1955, 1957, 1959, 1963, 1965]. Observing that a seemingly infinite variety of language was available to individual human beings based on clearly finite resources and experience, he proposed a formal representation of the rules or syntax of language, called generative grammar, that could provide finite--indeed, concise--characterizations of such infinite languages. Just as the breakthroughs in molecular biology in that era served to anchor genetic concepts in physical structures and opened up entirely novel experimental paradigms, so did Chomsky's insight serve to energize the field of linguistics, with putative correlates of cognitive processes that could for the first time be reasoned about 48 A

The prediction of protein structure from amino acid sequence has become the Holy Grail of computational molecular biology. Since Anfinsen [1973] first noted that the information necessary for protein folding resides completely within the primary structure, molecular biologists have been fascinated with the possibility of obtaining a complete three-dimensional picture of a protein by simply applying the proper algorithm to a known amino acid sequence. The development of rapid methods of DNA sequencing coupled with the straightforward translation of the genetic code into protein sequences has amplified the urgent need for automated methods of interpreting these one-dimensional, linear sequences in terms of three-dimensional structure and function. Although improvements in computational capabilities, the development of area detectors, and the widespread use of synchrotron radiation have reduced the amount of time necessary to determine a protein structure by X-ray crystallography, a crystal structure determination may still require one or more man-years.

He also taught the biochemistry course that I finally took, two years after finishing my Ph.D. David J. States deserves much of the credit as well. In the three years we have been working together, he greatly extended my understanding of not only what biologists know, but how they think. He has read several drafts of this chapter and made helpful suggestions. David Landsman, Mark Boguski, Kalí Tal and Jill Shirmer have also read the chapter and made suggestions. Angel Lee graciously supplied the gel used in Figure 4. Of course, all remaining mistakes are my responsibility.

An opposing school of thought, sometimes called "neural cy or "self-organizing systems,".

When making a decision a trust officer in a bank is confronted with a large assortment of information. In keeping with the postulates of this theory, the main postulates for the analysis of the investment decision process are that there exist: 1. A memory that contains lists of industries each of which has a list of companies associated with it. The memory also contains information associated with the general economy, industries, and individual companies. The set of rules constitutes the structure of the decision processes for an individual investor. It might be compared to the "rules of thumb" of the traditional "expert," but there is an important difference In common with other problem-solving programs, the processes are used iteratively and recursively. Lists of industries and companies are searched for particular attributes; sublists are created, searched and divided again. For example, to obtain a high growth portfolio, the list of companies stored in memory is searched to obtain securities with the desired pand) characteristics.

Baseball is a first step toward this goal.

How could one represent with a single prototype the class of figures which have an even number of disconnected parts?

This is an attempt to analyze attitudes and arguments brought forth by questions like "Can machines think?" and "Can machines exhibit intelligence?" Its purpose is to improve the climate which surrounds research in the field of machine or artificial intelligence. Its goal is not to convince those who answer the above questions negatively that they are ative wrong (although an attempt will be made to refute some of the neg arguments) but that they should be tolerant of research investigating these questions. Samuel Butler (1835-1902), in Erewhon and Erewhon Revisited (1933), concocted a civil war between the "machinists" and the "antimachinists." Butler stated "there is no security against the ultimate development of mechanical consciousness in the fact of machines possessing little consciousness now" and specylated that the time might come when "man shall become to the The topic came into prominence in the late 1940's when Babbage's dreams became a reality with the completion of the first large digital computers. When the popular press applied the term "giant brains" to these machines, computer builders and users, myself included, immediately arose to the defense of the human intellect. We hastened metic to proclaim that computers did not "think"; they only did arith A. M. Turing, who earlier had written one of the most important papers In it he circumvented the problem of properly defining the words "machine" and "thinking" and examined instead the question of a game This is now known throughout the computer field as "Turing's Test." Discussion of machine intelligence died down (but not out) in the early and mid-1950s but has come back in the last several years stronger than ever before. In fact, it has recently invaded the pages of Science (Mac-Gowan, 1960; Wiener, 1960; Taube, 1960; Samuel, 1960b). Like Turing, I avoid defining "to think." This notion is certainly not new, for it has existed since plicit man first compared his mental abilities with another man's, and it is im in all of the positive arguments on machine intelligence. Psychologists long ago developed "intelligence quotient" tinuum, as a yardstick in this con Existing commercial jet transports cannot transport people from one lake to another. But men cannot carry the load that a jeep can nor can men move with the speed of the jeep. Similarly, comparisons can be made between men and machines in the continuum of thinking. If there is objection to the use of the word "thinking," then "ability But it must be admitted that there exists some con of behavior in which men and machines coexist and in which they can be compared.

How could one represent with a single prototype the class of figures which have an even number of disconnected parts?