At the forefront of Artificial Intelligence

Artificial Intelligence Tour (Page 4)

Pioneers of AI: Prominest Research Scientists

The pioneers of AI are some of the brightest philosophers and scientists in this half century. Together, they defined the field of AI within only 50 years of time. The development of AI is considered the scientific wonder of the 20th century, parallel with the invention of Genetics Engineering.

Marvin Minsky

Marvin Minsky, considered by many to be the father of AI, has made major contributions to nearly every discipline in artificial intelligence. While he is also the founder of the MIT AI labs, he is mostly well known for his book The Society of the Mind. In his undergraduate years, he developed interests in many different fields (physics, psychology, neurology, and mathematics). Combining knowledge from these many fields, he was able to develop ingenious computational ideas to characterize human psychological processes; influenced by Skinner's behaviorism, Minsky, together with Dean Edmond, built SNARC in 1951 to simulate the learning of mouse in walking through a maze. Composing of vacuum tubes, SNARC was the first neural network ever built in the history. During 1950's, Minsky and some other earliest AI scientists continued building a groundwork for AI. The process reached its peak at the Dartmouth Conference of 1956. In 1959, Minsky, together with John McCarthy, founded the MIT AI Lab and officially started his research on AI. His seminal papers in 1961, "Steps Towards Artificial Intelligence", and in 1963, "Matter, Mind, and Models", proposed a general directions for AI researchers. In 1969, Minsky, together with Seymour Papert, published the book "Perceptrons" on the potential of a loop-free learning and pattern recognition machines. During 1970's, Minsky and Papert began formulating the theory called "The Society of Mind" which proposed that the human mind is not composed of a few basic mechanisms, but an interaction of a diverse variety of resourceful agents. In "A Framework for Representing Knowledge" (1974), Minsky presented a model of knowledge representation. These representations, called "frames", inherit their value from previously defined frames. The structure of such presentation is often considered as the early form of object oriented programming. In 1985, Minsky organized a collection of essays he had been working on since 1970's and published "The Society of Mind". Minsky is currently working on his next book, "The Emotion Machine", which addresses the motivations that drive the agents in the society of mind.

Roger Schank

Roger Schank has dedicated his life on Natural Language and Cognitive Science - The study of learning and perception. His revolutionary insight on language pushed the study of natural language one further step in 1970's. Unlike conventional approach of defining the meaning of each word, Schank proposed a scheme that reduced all English verbs into eleven basic acts. This scheme stressed on the meaning of the language instead of the wording of the language. In 1973, MARGIE, constructed by Schank's students, employed Schank's sematic scheme to generate plausible references from a sentence. Even though MARGIE turned out to be inefficient in deriving the essential idea, it was the first embodiment of conceptual dependency theory. To improve the original scheme, Schank introduced the concept of "script" which is a general plan for a series of actions. SAM was then built according to the script theory to interpret articals on car accidents. SAM turned out to be a great success! In fact, SAM became the first machine translator ever. With an understanding of the true meaning of the sentence, SAM wasn't limited to a certain language and could paraphrase its understanding in different languges. Schank then dedicated himself in Cognitive Science. In 1989, he founded The Institute for the Learning Sciences in Northwestern Univeristy. He is the author of over a dozen books including "Scripts, Plans, Goals and Understanding", "The Cognitive Computer: On Language Learning and Artificial Intelligence", and "The Connoisseur's Guide to the Mind".

Joseph Weizenbaum

Joseph Weizenbaum became involved in the AI research in the late 1950's. Like Roger Schank, he contributed greatly to the study of Natural Language. His major accomplishments are SLIP, a list-processing language, and ELIZA, the first computer psychiatrist. Weizenbaum holds an unique attitude toward the ethics problems of AI research. Unlike other AI scientists who tend to put the technical advance above the ethics, Weizenbaum pays no less attention on the moral aspect of AI than on the research itself. In his book- "Computer Power and Human Reason", Weizenbaum regards computers merely as tools to expedite our daily lives. Even though he invented ELIZA which has the ability to talk almost like a human, Weizenbaum never believed that machines should ever substitute humans in any way. Weizenbaum's philosophy appeals for a deeper understanding of the purpose of AI research.

Allen Newell

Allen Newell contributed greatly to the early development of AI. With Marvin Minsky, John McCarthy, and Herbert Simon, Newell helped defined the field of AI in the Dartmouth Conference in 1956. His study on heuristic reasoning earned him a position in RAND Corp. In there, Newell and Simon developed the Logic Theorist which is considered the first AI program in the history. Logic Theorist was able to solve intriguing math problems with its heuristic tree scheme. Newell and Simon then moved to refine the Cognitive thoery in the following years. Researchers in 1970's have abandoned their hope in defining intelligence with a few basic rules and moved to the interactive theories which propose intelligence as a result of the interaction between million pieces of common sense. The change in research direction demands for new theories and projects. With their study in Cognitive Science in 1960's, Newell and Simon came up with the theory on Production System and the Soar project. Soar's ability to refine its goal and evaluate the possible outsomes made itself the precursor of AI gaming (chess, tictactoe, checkers) programs. After his death in 1992, Allen Newell was both remembered for his contribution to science and his impressive characters.

Craig Reynolds

For years, almost all I did was hack on computer animation. All my friends said "Craig, get a life!" I thought they said alife and so began to study the field of Artificial Life. - Craig Reynolds

Craig Reynolds is considered to be one of the major pioneers in the ever expanding field of artificial life. He was the first to implement Alife (artificial life) to animations involving large numbers of objects. His model of flocking behavior in animals greatly aided animators who were faced with the problem of animating large crowds. This model, is otherwise known as 'boids' for which he is internationally renowned for. Boids works through requiring each individual member of a flock to follow three simple rules (don't fall behind, keep up with nearby boids, try to stay a minumum distance between your neighbors and obstacles, move towards what seems to be the center of mass of nearby boids). It is quite phenomenal that the result of something so simple could have brought something so life-like as a real flock of birds.

...I realized that it must be a very different experience to be a member of a flock than to observe it from the outside, much like the difference between driving in traffic and standing on a roadside watching traffic whiz by. So I just tried to mentally put myself inside a flock and imagine what I would have to do to fly with them. I'd have to make sure I didn't get too close to any of my local flockmates. I'd want to be flying at the same speed and in the same heading as my local flockmates. (This also means I'm unlikely to collide with them in the near future.) And finally, if I noticed that all of my local flockmates where on one side of me, I'd want to drift over towards them. It seemed that these three rules would be necessary conditions for flocking, but I didn't know if they would be sufficient. But once I tried the experiment (in 1986) by implementing the three rule model, it was clear that those three rules sufficed. - Craig Reynolds (Generation 5 Interview)

Craig Reynolds was involved with over seven movies (including Disney's TRON and Warner Brothers' Batman Returns). In 1998, he recieved the Scientific And Engineering Award presented by The Academy of Motion Picture Arts and Sciences for his contributions to the development of three-dimensional computer animation. Craig Reynold's boids page can be found at http://hmt.com/cwr/boids.html.

Generation 5 Interview with Marvin Minsky

Generation 5 Interview with Roger Schank

Generation 5 Interview with Craig Reynolds

Submitted: 10/12/1999

Article Toolbar

BibTeX entry