The general strategy of Meta-DENDRAL is to reason from data to plausible generalizations and then to organize the generalizations into a unified theory. Three main subprobleras are discussed: (1) explain the experimental data for each individual chemical structure, (2) generalize the results from each structure to all structures, and (3) organize the generalizations into a unified theory. The program is built upon the concepts and programmed routines already available in the Heuristic DENDRAL performance program, but goes beyond the performance program in attempting to formulate the theory which the performance program will use. I. Introduction Theory formation in science embodies many elements of creativity which make it both an interesting and challenging task for artificial intelligence research. One of the goals of the Heuristic DENDRAL project has long been the study of processes underlying theory formation.

This edge follower could easily find the outlines of white cubes on a black table, but was prone to error in less carefully controlled environments. Our studies of its inadequacies have stimulated the development of a more powerful edge follower, which overcomes most of the limitations of the old one. This program is currently the initial stage of visual processing in the Stanford hand-eye system (2). It has demonstrated an ability to track weak edges under adverse lighting conditions 2. HARDWARE The edge follower uses a standard vidicon television camera, modified to provide computer control of orientation (a pan-tilt head), focal length (a lens turret), color filter, focus, and target voltage. The lens iris is set manually. The pan-tilt head, lens turret, and focus motor *This research was supported by the Advanced research Projects Agency of the Department of Defense under Contract No. SD-183. The views and conclusions contained in this document are those of the authors and should not be interpreted as necessarily representing the official policies, either expressed or implied, of the Advanced Research Projects Agency of the U.S. Government.

This system accepts as input a scene represented as a line drawing. Based on a set of known object models, the program attempts to determine the identity and location of each object viewed. The most significant feature of the system is its ability to deal with imperfect input data. This ability appears essential in light of our current stock of preprocessing techniques and the variation that is possible in real world data. INTRODUCTION A hand-eye system is a problem solving system with an eye (camera) for input and a hand (manipulator) for output. Such a system must have at least 1) a set of scene analysis (perception) programs which interpret the real world in a meaningful way, 2) a set of manipulation programs which control movement of the hand in 3-space, and 3) an executive (problem solver, strategy) program which directs the perceptual and motor processes toward a desired goal.

In R. J. O'Brien (Ed.), Report of the twenty-second annual round table meeting on linguistics and language studies. (Monograph Series on Languages and Linguistics, No. 24.) Washington, D.C.: Georgetown University Press, 35-56.

"Heuristic DENDRAL is a computer program written to solve problems of inductive inference in organic chemistry. This paper will use the design of Heuristic DENDRAL and its performance on different problems for a discussion of the following topics: 1. the design for generality; 2. the performance problems attendant upon too much generality; 3. the coupling of expertise to the general problem solving processes; 4. the symbiotic relationship between generality and expertness, and the implications of this symbiosis for the study and design of problem solving systems. We conclude the paper with a view of the design for a general problem solver that is a variant of the "big switch" theory of generality."See also: Stanford University Report (ACM Citation)In Meltzer, B. and Michie, D. (Eds.), Machine Intelligence 6, pp. 165–190. Edinburgh University Press

"Problem-solving methods using some sort of heurstically guided search process have been the subject of much research in Artificial Intelligence. This paper groups these problem-solving methods under three major headings: the State-Space Approach, the Problem-Reduction Approach and the Formal-Logic Approach." New York: McGraw-Hill.

The problem we consider in this paper is that of discovering formal rules which will enable us to decide when a question posed in English can be answered on the basis of one or more declarative English sentences. To illustrate how this may be done in very simple cases we give rules which translate certain declarative sentences and questions involving the quantifiers'some', 'every', 'any', and'no' into a modified first-order predicate calculus, and answer the questions by comparing their translated forms with those of the declaratives. We suggest that in order to capture the meanings of more complex sentences it will be necessary to go beyond the first-order predicate calculus, to a notation in which the scope of words other than quantifiers and negations is clearly indicated. We conclude by describing a notational form for connected sentences, which seems to be a natural extension of Chomsky's'deep structures'.

We describe the analysis of visual scenes consisting of black on white drawings formed with curved lines, depicting familiar objects and forms: houses, trees, persons, and so on; for instance, drawings found in coloring books. The goal of such analysis is to recognize (by computer) such forms and shapes when present in the input scene; that is, to name (correctly) as many parts of the scene as possible: finger, hand, girl, dance, and so on. Complications occur because each input scene contains several such objects, partially occluding each other and in varying degrees of orientation, size, and so on. The analysis of these line drawings is an instance of'the context problem', which can be stated as'given that a set (a scene) is formed by components that locally (by their shape) are ambiguous, because each shape allows a component to have one of several possible values (a circle can be sun, ball, eye, hole) or meanings, can we make use of context information stated in the form of models, in order to single out for each component a value in such manner that the whole set (scene) is consistent or makes global sense?' Thus, shape drastically limits the values that a component could have, and further disambiguation is possible only by using global information (derived from several components and their interrelations or interconnections) under the assumption that the scene as a whole is meaningful. This paper proposes a way to solve'the context problem' in the paradigm of coloring book drawings. We have not implemented this approach; indeed, a purpose of this paper is to collect criticisms and suggestions.

One example of this will suffice.

We have attempted to discover formal rules for transcribing into musical notation the fugue subjects of the Well-Tempered Clavier, as this might be done by an amanuensis listening to a'deadpan' performance on the keyboard. In this endeavour two kinds of problem arise: what are the harmonic relations between the notes, and what are the metrical units into which they are grouped? The harmonic problem is that the number of keyboard semitones between two notes does not define-- their harmonic relation, and we further develop an earlier theory of such relations, arriving at an algorithm which assigns every fugue to the right key and correctly notates every accidental in its subject.