Page images
PDF
EPUB

ARTIFICIAL INTELLIGENCE

Nils J. NILSON
Artificial Intelligence Center, Stanford Research Institute

Menlo Park, California 94025, USA

(INVITED PAPER)

This paper is survey of Artificial Intelligence (AI). It divides the field into four core topics (ambodying the base for a science of intelligence) and eight applications topics (in which research has been contributing to core ideas). The paper discusses the history, the major landmarks, and some of the controversies in each of these twelve topics. Each topic is represented by a chart citing the major references. The se references are contained in an extensive bibliography. The paper concludes with a discussion of some of the criticisms of AI and with some predictions about the course of future research.

1.

INTRODUCTION

Can we ever hope to understand the nature of intelli-
gence in the same sense that we understand, say, the
nature of flight? Will our understanding of intel-
ligence ever be sufficient to help us build working
models--machines that think and perceive--in the same
way that our understanding of aerodynamics helps us
build airplanes? Intelligence seems so varied. We
see it when a chemist discovers the structure of a
complex molecule, when a computer plays chess, when
i na thematician finds a proof, and even when a child
walks home from school. Are there basic mechanisms
or processes that are common to all, of these activi-
ties and to all others commonly thought to require
intelligence?

way in which he views himself. In this paper, I
vill try to describe the Al campaign, how it seems
to be organized into subcampaigns, who is doing
what, some of the current internal controversies,
and the main achievements. There is the usual word
of caution: I've made some rather large simplifica-
tions in attempting to stand aside from the field
and look at it with perspective. Not all workers
would necessarily agree with what follows.

'The field of Artificial Intelligence (AI) has as its
main tenet that there are indeed common processes
that underlie thinking and perceiving, and further-
more that these processes can be understood and
studied scientifically. The processes themselves do
not depend on whether the subject being thought about
or perceived is chemistry, chess, mathematics, or
childhood navigation. In addition, it is completely
unimportant to the theory of Al who is doing the
thinking or perceiving--man or computer. This is an
implementational de tail.

Before beginning we must discuss an important char-
acteristic of AI as a field, namely, that it does
not long retain within it any of its successful ap-
plications. Computer aides to mathematicians, such
as differential equation solvers, that originated
(at least partly) from Al research, ultimately be-
come part of applied mathematics. A system, named
DENDRAL, that hypothesizes chemical structures of
organic molecules based on mass-spectrogram data is
slowly escaping its Al birthplace and will likely
become one of the standard tools of chemists. This
phenomenon is well-recognized by Al researchers and
has led one of them to state that Al is known as the
"no-win" field. It exports all of its winning ideas.

On reflection, this is not surprising. When a field
takes as its subject matter all of thinking, and
then when particular brands of that thinking are
applied to chemistry, nathematics, physics, or what-
ever, these applications become parts of chemistry,
mathematics, physics, etc. When people think about
chemistry, we call it part of chemistry--not an ap-
plication of psychology. The more successful Al be-
comes, the more its applications will become part of
the application area,

These are the energing beliefs of a group of computer
scientists claiming to be founding a new science of
Intelligence. While attempting to discover and
understand the basic nechanisms of intelligence,
these researchers have produced working models in the
form of computer programs capable of some rather im-
pressive feats: playing competent chess, engaging
in limited dialogs with humans in English, proving
reasonably difficult nathematical theorems in set
theory, analysis, and topology, guessing (correctly)
the structure of complex organic molecules from mass-
spectrogran data, assembling mechanical equipment
with a robot hand, and proving the correctness of
small computer programs.

Destined apparently to lack an applied branch, is
there a central core or basic science of Al that
will continue to grow and contribute needed ideas to
applications in other areas? I think the answer is
yes. Just what form these central ideas will ulti-
mately take is difficult to discern now. Will Al be
something like biology--diverse but still united by
the common structure of DNA? What will be the DNA
of AI?

1

Whether the activities of these workers constitute a
new scientific field or not, at the very least Al 18
a major campaign to produce some truly remarkable
computer abilities. Like going to the moon or
creating 118e, it is one of man's grandest enter-
prises. As with all mand enterprises, it will have
profound influences on nan's way of life and on the

Or will the science of Al be more 11ke the whole of
science itself--united by little more than sone
vague general principles such as the scientific
method? It is probably too early to tell. The

presont central ideas sau more specific than does the scientific method but less concrete than DNA.

the applications areas themselves. Until all of the principles of intelligence are uncovered, AI researchers will continue to search for then in various first-level applications areas.

2.

WHAT IS HAPPENING IN AI?

2.1 The structure of the field

As a tactic in attempting to discover the basic principles of intelligence, Al researchers have set themselves the preliainary goal of building computer progrus that can perfon various intellectual tasks that humans can perfon. There are major projects currently under way whose goals are to understand natural language (both written and spoken), play naster chess, prove non- trivial nathematical theoreus, vite. computer programs, and so forth. These projects serve two purposes. First, they provide the appropriate settings in which the basic mechanisas of intelligence can be discovered and clarified. Second, they provide non-trivial opportunities for the application and testing of such mechanisms that are already known. I am calling these projects the first-level applications of Al.

Figure 1, then, divides work in Al into twelva major topics. I have attempted to show the major papers, projects, and results in each of these topics in Charts 1 through 12, each containing references to an extensive bibliography at the end of this paper. These charts help organize the literature as well as indicate some thing about the structure of work in the field. By arrows linking boxes within the charts we attempt to indicate how work has built on (or has been provoked by) previous work. The items in the bibliography are coded to indicate the subheading to which they belong. I think that the charts (taken as a whole) fairly represent the important work even though there may be many differences of opinion among workers about some of the entries* (and especially about how work has built on previous work).

Obviously, a short paper cannot be exhaustive. But in this section I will summarize what is going on in AI research by discussing the major accomplishments and status of research in each of the twelve subheadings.

I have grouped these first-level applications (somewhat arbitrarily) into eight topics shown spread along the periphery of Figure 1. These are the eight that I think have contributed the most to our basic understanding of intelligence. Each has strong ties to other (non-AI) fields, as well as to each other; the major external ties are indicated by arrows in Figure 1.

2.2 The core topics

Basić mechanisms of intelligence and implementational techniques that are common to several applications, I call core topics, It seems to me that there are four major parts to this central core:

Fundamentally, AI is the science of knowledge--how to represent knowledge and how to obtain and use knowledge. Our core topics deal with

these fundamentals. The four topics are highly interdependent, and the reader should be warned that it is probably wrong to attempt to think of them separately even though we are forced to write about them separately.

2.2.1 Common-sense reasoning, deduction, and

problem-solving (Chart 1)

• Techniques for modeling and representation of

knowledge. Techniques for common sense reasoning, deduction,

and problem solving. • Techniques for heuristic search. • AI systems and languages.

These four parts are shown at the center of Figure 1. Again, we have indicated ties to other fields by arrows. It must be stressed that most AI research takes place in the first-level applications areas even though the primary goal may be to contribute to the more abstract core topics.

By reasoning, etc., we mean the major processes involved in using knowledge: Using it to make inferences and predictions, to make plans, to answer questions, and to obtain additional knowledge. As a core topic, we are concerned mainly with reasoning about everyday, common domains (hence, common sense) because such reasoning is fundamental, and we want also to avoid the possible trap of developing techniques applicable only to some specialized domain, Nevertheless, contributions to our ideas about the use of knowledge have come from all of the applications areas.

There have been three major themes evident in this core topic. We night label these puzzle-solving, question-answering, and common-sense reasoning.

If an application is particularly successful, it Bight be noticed by specialists in the application area and developed by then as a useful and economically viable product. Such applications we might call second-level applications to distinguish them from the first-level applications projects undertaken by the Al researchers themselves. Thus, when Al researchers work on a project to develop a prototype system to understand speech, I call it a firstlevel application. II General Motors were to develop and install in their assembly plants a system to interpret television inages of automobile parts on a conveyor belt, I would call it a secondlevel application. (We should humbly note that perhaps several second-level applications will emerge without benefit of obvious Al parentage. In fact, these may contribute nightily to Al science itsell.)

Puzzle-solving. Early work on reasoning concentrated on writing computer prograns that could solve simple puzzles (tower of Hanoi, missionaries and cannibals, logic problems, etc.). The Logic Theorist and GPS (see Chart 1) are typical examples. From this work certain problem-solving concepts were developed and clarified in an uncluttered atmosphere. Among these were the concepts of heuristic search, problem spaces and states, operators (that transformed one problem state into ano ther), goal and subgoal states, meansends analysis, 'and reasoning backwards. The fact

Thus, even though I agree that Al is a field that cannot retain its applications, it is the secondlevel applications that it lacks. These belong to

In particular, some might reasonably clain machine vision (or more generally, perception) and language understanding to be core topics,

[blocks in formation]

FIGURE I MAJOR SUB-PARTS OF AI SHOWING TIES TO OTHER FIELDS

[merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][ocr errors][merged small][merged small]
« PreviousContinue »