I. What is artificial intelligence?
There's no single standard definition of the term "artificial
intelligence" we'll see if we can't prompt you to construct
one of your own through looking at definitions offered (mostly) by
several seminal textbooks. But first, we should note that the name
"artificial intelligence" itself does not have universal support
from the people who work in this field. It's certainly the most
commonly used name, but you'll also hear and see "computational
intelligence" (which is the name used here at UBC), "machine
intelligence", and "intelligent systems" (the current favourite
at my previous institution). Regardless of the name used, one
gets the sense, just from the names, that this endeavour has
something to do with intelligence and computers or other man-made
artifacts. (And please note that I've used two "u" words just in
this paragraph. This Canadian English thing is gonna be easy. Now
if someone could just explain the "zed" thing....)
For more insight into what artificial intelligence, or AI, is, we
ask the experts:
Artificial intelligence is the science of making machines do things
that would require intelligence if done by men.
Marvin Minsky (1968)
Marvin Minsky is regarded as one of the founders of the field of AI,
and I'm sure that his use of "men" was intended in the "scientific"
sense ("man" as a species distinction as opposed to a gender
distinction) and wasn't intended to slight women. Still, it was
the 60s, and you'd think Marvin's consciousness would have been
raised. But I digress. His definition, while succint, avoids the
issue of defining intelligence. It's a recurring theme.
[AI researchers] try to get machines to exhibit behavior that we
call intelligent behavior when we observe it in human beings.
Since the machine is almost always a computer, Artificial
Intelligence is a branch of computer science.
James Slagle (1971)
Slagle's definition isn't much different from Minsky's, but it's
included here because he explains why AI is part of computer science.
He also goes out of his way to avoid confusion and outrage on the
part of his reader by using the term "human beings", thereby including
the more intelligent gender along with the couch-potato gender
(says your intrepid instructor as he types this on his laptop while
reclining on his couch).
Artificial Intelligence is not the study of computers, but of
intelligence in thought and action. Computers are its tools,
because its theories are expressed as computer programs that
enable machines to do things that would require intelligence
if done by people...
Margaret Boden (1977)
Maggie Boden is a cognitive scientist, as opposed to a computer
scientist, who writes eloquently about artificial intelligence.
Not being a computer nerd, she makes explicit the point that
AI involves computer programs...something that computer nerds
seem to think everyone knows already.
Artificial Intelligence is the study of ideas that enable computers
to be intelligent. But what is intelligence? Is it the ability to
reason? Is it the ability to acquire and apply knowledge?. Is it
the ability to perceive and manipulate things in the physical world?
Surely all of these abilites are part of what intelligence is, but they
are not the whole of what can be said. A definition in the usual
sense seems impossible because intelligence appears to be an
amalgam of so many information-representation and information
processing talents.
Nevertheless, the goals of the field of Artificial Intelligence can
be defined as follows:
One central goal of Artificial Intelligence is to make computers
more useful.
Another central goal is to understand the principles that make
intelligence possible.
Patrick Winston (1984)
There's more here than what we were able to fit on the PowerPoint
slide in class. The essence, though, is that Winston acknowledges
that even AI people have difficulty grappling with the word
"intelligence". Still, we press on...
Artificial Intelligence is the study of mental faculties through
the use of computational models....The fundamental working
assumption, or "central dogma" of AI is this: What the brain does
may be thought of at some level as a kind of computation.
Eugene Charniak and Drew McDermott (1985)
Charniak and McDermott, on the other hand, emphasize the scientific
aspects of AI over the engineering (i.e., programming) aspects, but
they include one of the tenets of AI work: whatever intelligence
is, it can be viewed as some form of computation. Since computation
happens on platforms other than brains, it's not too much of a stretch
to consider that intelligence could happen on platforms other than brains,
no?
Artificial intelligence attempts to understand intelligent entities.
Thus, one reason to study it is to learn more about ourselves. But
unlike philosophy and psychology, which are also concerned with
intelligence, AI strives to build intelligent entities...[which]
are interesting and useful in their own right.
Stuart Russell and Peter Norvig (1995)
There's nothing much new here. Russell and Norvig reinforce the
notion that there's a scientific side and an engineering side to AI.
That message comes through more clearly in the definition provided
by the authors of our textbook, though they use the name "computational
intellligence":
Computational intelligence is the study of the design of intelligent
agents. An agent is something that acts in an environment -- it does
something.... An intelligent agent is a system that acts intelligently:
What it does is appropriate for its circumstances and its goal....
The central scientific goal of computational intelligence is to understand
the principles that make intelligent behavior possible.... The main
hypothesis is that reasoning is computation. The central engineering
goal is to specify methods for the design of useful, intelligent
artifacts.
David Poole, Alan Mackworth, and Randy Goebel (1998)
Perhaps the most succinct definition is this:
Artificial intelligence is the science of making computers behave the
way they do in the movies.
unknown
More next time....
Last revised: September 13, 2004