In this post we are going to discuss history of the artificial intelligence. We can divide this history into the following time period. One point of caution that these periodor not disjoint but overlapping based on the idea.
1. The gestation period of AI (1943-1955)
The first work that is now generally recognized as AI was done by Warren McCulloch and Walter Pitts (1943). They proposed a model of artificial neurons in which each neuron is characterized as being ON or OFF. They showed that any computable function could be computed by some network of connected neurons. They also suggest that suitably defined neural networks could learn.
Donald Hebb (1949) demonstrated a simple updating rules for modifying the connection strength between neurons. His rule now called Hebbian Learning, remains an influential model till today.
Two undergraduate student at Harvard, Marvin Minsky and Dean Edmonds built the first neural network computer called SNARC in 1950. Minsky later prove an influential theorems showing the limitations of neural network research.
There were a number of early examples of work that can be characterized as AI, but Alen Turing’s ,vision was perhaps the most influential. In his 1950’s article “Computing Machinery and Intelligence“, he introduced the Turing Test, Machine Learning, Genetic Algorithm and Reinforcement Learning. He proposed the child programme idea, explaining “Instead of trying to produce a programme to simulate the adult mind, why not rather try to produce one which simulate the child’s“.
2. The birth of Artificial Intelligence (1956)
Princeton was the home of another influential figure in AI, John McCarthy. After receiving his PhD there in 1951 and working for two years as an instructor, McCarthy moved to Stanford and then to Dartmouth College, which was to become the official birthplace of the field.
McCarthy convinced Minsky, Claude Shannon and Nathaniel Rochester to help him bring together US researcher interested in automata theory, neural nets and the study of intelligence. They organized two month workshop at Dartmouth College in the summer of 1956. The proposal states that:
We propose that a 2 month, 10 man study of artificial intelligence be carried out during the summer of 1956 at Dartmouth College in Hanover, New Hampshire. The study is to proceed on the basis of the conjecture that every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it……..
There are total 10 attendees in all, including Trenchard More from Princeton, Arthur Samuel from IBM and Ray Solomonoff and Oliver Selfridge from MIT.
Two researchers from Carnegie Tech (Now Carnegie Mellon University) Allen Newell and Herbert Simon presented a reasoning program, the Logic Theorist(LT), about which simon claimed, “We have invented a computer program capable of thinking non-numerically, and thereby solved the venerable mind-body problem.“. Soon after the workshop, the program was able to prove most of the theorem in chapter 2 of Russel and whitehead’s Principia Mathematica.
The Dartmouth workshop did not lead to any new breakthrough, but it did introduce all the major figures to each other. For the next 20 years, the field would be dominated by these people and their students and colleague at MIT, CMU, Stanford and IBM.
3. Early enthusiasm, great expectations (1952-69)
The early years of AI were full of success—in a limited way. Given the primitive computers and programming tools of that time and the fact that only a few year earlier computer were seen as things that can do arithmetic and no more, it was astonishing whenever a computer do anything remotely clever.
Newell and simmon’s early success was followed up with the General Problem Solver (GPS). This program was designed from start ti imitate human-problem solving protocol. Herbert Gelernter (1959) constructed the Geometry Theorem Prover which able to prove theorems that many student of mathematics would find quite tricky.
In 1952, Arthur Samuel wrote a series of program for checkers that eventually learned to play at a strong amateur level. Along the way he disproves the idea that computers can do only what they are told to do: his program quickly learned to play a better game than its creator.
John McCarthy moved from Dartmouth to MIT and there made three crucial contribution in one historic year 1958:
- McCarthy defined the high level programming language LISP, which become dominant programming language for the next 30 years.
- With LISP, McCarthy had the tool he needed, but access to scarce and expensive computing resources was also a serious problem. In response, he and other at MIT invented time sharing system.
- Published a paper entitled Program with common sense, in which he describe the Advice Taker, a hypothetical program that can be seen as the first complete AI system. Like the Logic Theorist and Geometry Theorem Prover, McCarthy’s program was designed to use knowledge to search for solutions to problems. But unlike the others, it was embodied general knowledge of the world.
Minsky supervised a series of student who choose limited problems that appeared to require intelligence to solve. These limited domains became known as microworld.
- James Slagle’s SAINT program(1963) was able to solve closed form calculus integration problems typical of first year college calculus.
- Tom Evan’s ANALOGY program(1968) solve geometric analogy problem that appears in IQ test.
- Daniel Bobrow’s STUDENT program(1967) solved algebra story problem such as the following:
If the number of customer Tom gets is twice the square of 20 percent of the number of advertisement he runs, and the number of advertisements he runs is 45, what is the number of customers Tom gets?
Early work on building the neural network also flourished. Hebbs learning method were enhance by Bernie Widrow(1960) who called his network adalines, and by Frank Rosenblatt(1962) with his perceptrons.
4. A dose of reality (1966-73)
From the beginning, AI researchers were not shy about making predictions of their coming success. Simon predicted that: “Within 10 years a computer would be chess champion, and a significant mathematical theorem would be proved by machine” These prediction came true (or approximately true) within 40 years rather than 10. Simon’s overconfidence was due to the promising performance of early AI systems on complex examples. In almost all cases, however, these early systems turned out to fail miserably when tried out on wider selections of problems and on more difficult problem.
- The first kind of difficulty arose because most early programs knew nothing of their subject matter; they succeeds by means of simple syntactic manipulations. A typical story occurred in early machine translation effort, which was generously funded by the US National Research Council in an attempt to speed up the Russian scientific papers. It was thought initially that simple syntactic transformation based on the grammars of Russian and English, the word replacement from an electronic dictionary, would suffice to preserve the exact meaning of the sentences. The fact that accurate translation requires background knowledge in order to resolve ambiguity and establish the content of the sentence. The famous re-translation of “the spirit is willing but the flesh is weak” as the “vodka is good but the meat is rotten” illustrates the difficulties encountered. In 1966, a report by an advisory committee found that “there has been no machine translation of general scientific text, and none is in immediate prospect.” All US government funding for academic translation projects was canceled.
- The second kind of difficulty was the intractability of many of the problems that AI was attempting to solve.
- A third difficulty arose because of some fundamental limitations on the basic structures being used to generate intelligent behavior. For example, Minsky and Papert’s book Perceptrons(1969) proved that, although perceptrons could be shown to learn anything they were capable of representing, they could represent very little. In particular, a two input perceptron could not be trained to recognize when its two inputs were different. Although their result do not apply to more complex, multilayer networks, research funding for neural-net research soon dwindled to almost nothing.
5. Knowledge-based System: Era of (Expert System)? (1969-79)
The picture of problem solving that had arisen during the first decade of AI research was of general purpose search mechanism trying to string together elementary reasoning steps find the complete solutions. Such approaches have been called weak method because, although, general they do not scale up to large or difficult problem instances. The alternative to weak method is to use more powerful, domain specific knowledge, that allows larger reasoning steps and can more easily handle typically occurring cases in narrow areas of expertise.
The DENDRAL program(1969) is an example of this approach. It infer the molecular structure from the information provided by the mass spectrometer.
The next major breakthrough is the program MYCIN(1970) which is able to diagnose blood infections. With about 450 rules MYCIN was able to perform considerably better than junior doctors.
6. AI becomes an industry(1980-present)
The first successful commercial expert system, R 1, began operation at the Digital Equipment Corporation(1982). The program help to configure orders for new computer systems, by 1986, it was saving the company an estimated $40 million a year. By 1988, DEC’s AI group has 40 Expert system deployed, with more on the way.
In 1981, Japan announced the “Fifth generation” project a 10 years plan to build intelligent computers running Prolog. In response, the United States formed the Microelectronics and Computer technology Corporation(MCC) as a research consortium designed to assure national competitiveness.
Overall, the AI industry boomed from a few million dollars in 1980 to billions of dollars, in 1988, including hundreds of companies building expert systems, vision systems, robots and software and hardware specialized for these purposes.
Soon after that came a period called the AI winter, in which many companies fell by the wayside as they failed to deliver on extravagant promises.
7. The return of Neural Network(1986-present)
In the mid 1980 at least four different groups reinvented the the back-propagation learning algorithm first found in 1969 by Bryson and Ho. The algorithm is applied to many learning problems in computer science and psychology, and the widespread dissemination of the result in the collection Parallel Distributed Processing (1986) caused great excitement.
With increase in computational power and availability of large datasets neural network achieve tremendous success in the field of speech recognition, computer vision and game playing.
8. AI adopt the Scientific Method(1987-present)
Recent years have seen a revolution in both the content and the methodology of work in artificial intelligence. It is more common to build on existing theories than to propose brand new ones. In terms of methodology artificial intelligence finally comes firmly under the scientific method. To be accepted, hypotheses must be subjected to rigorous empirical experiment, and the result must be analyzed statistically for their importance. It is now possible to replicate experiments by shared repositories of test data and code.
Field of speech recognition, embraced Hidden Markov Models (HMM). Two aspects of HMM are relevant:
- They are based on rigorous mathematical theory. This has allowed speech researcher to build on several decades of mathematical result developed in other field.
- They are generated by a process of training on a large corpus of real speech data. This ensures that the performance is robust.
Speech technology and the related field of handwritten character recognition are already making the transition to wide spread industrial and consumer application.
Machine Translation followed the same course as speech recognition. In the 1950 there was initial enthusiasm for an approach based on sequence of words, with models learned according to the principles of information theory. That approach fell out of favor in the 1960, but returned in the late 1990s and now dominate the field.
9. Emergence of Intelligent Agent(1995-present)
Perhaps encouraged by the progress in solving the sub problems of AI, researchers have also started to look at the whole agent problem again. The work of Allen Newell, John Laird and Paul Rosenbloom on SOAR is the best known example of a complete agent architecture.
One of the most important environments for intelligent agents is the Internet. Search engines, recommender system and web site aggregators all uses the AI.
Recent progress in the control of robotic cars has derived from a mixture of approaches ranging from better sensors, control-theoretic integration of sensing , localization and mapping, as well as a degree of high level planning.
In 1995 Richard Wallace develops the chatbot A.L.I.C.E (Artificial Linguistic Internet Computer Entity), inspired by Joseph Weizenbaum’s ELIZA program, but with the addition of natural language sample data collection on an unprecedented scale, enabled by the advent of the Web.
In 1997 Deep Blue becomes the first computer chess-playing program to beat a reigning world chess champion.
In 2000 Honda’s ASIMO robot, an artificially intelligent humanoid robot, is able to walk as fast as a human, delivering trays to customers in a restaurant setting.
7. The availability of large data sets(2001-present)
In 2004, The first DARPA Grand Challenge, a prize competition for autonomous vehicles, is held in the Mojave Desert. None of the autonomous vehicles finished the 150-mile route.
In 2006, Oren Etzioni, Michele Banko, and Michael Cafarella coin the term “machine reading,” defining it as an inherently unsupervised “autonomous understanding of text.” In the same year Geoffrey Hinton publishes “Learning Multiple Layers of Representation,” summarizing the ideas that have led to “multilayer neural networks that contain top-down connections and training them to generate sensory data rather than to classify it,” i.e., the new approaches to deep learning.
In 2007, Fei Fei Li and colleagues at Princeton University start to assemble ImageNet, a large database of annotated images designed to aid in visual object recognition software research.
In 2009, Google starts developing, in secret, a driverless car. In 2014, it became the first to pass, in Nevada, a U.S. state self-driving test.
In 2009, Computer scientists at the Intelligent Information Laboratory at Northwestern University develop Stats Monkey, a program that writes sport news stories without human intervention.
In 2010, Launch of the ImageNet Large Scale Visual Recognition Challenge (ILSVCR), an annual AI object recognition competition.
In 2011, A convolutional neural network wins the German Traffic Sign Recognition competition with 99.46% accuracy (vs. humans at 99.22%). In same year Apple’s Siri, Google’s Google Now and Microsoft’s Cortana are smartphone apps that use natural language to answer questions, make recommendations and perform actions.
In 2011 Watson, a natural language question answering computer, competes on Jeopardy! and defeats two former champions. In the same year Researchers at the IDSIA in Switzerland report a 0.27% error rate in handwriting recognition using convolutional neural networks, a significant improvement over the 0.35%-0.40% error rate in previous years.
In October 2012, A convolutional neural network designed by researchers at the University of Toronto achieve an error rate of only 16% in the ImageNet Large Scale Visual Recognition Challenge, a significant improvement over the 25% error rate achieved by the best entry the year before.
In March 2016 Google DeepMind’s AlphaGo defeats Go champion Lee Sedol.
- Artificial Intelligence: A Modern Approach by Stuart Russell and Peter Norvig
- Time Line of Artificial Intelligence
This article is contributed by Ram Kripal. If you like eLgo Academy and would like to contribute, you can mail your article to email@example.com. See your article appearing on the eLgo Academy page and help other Geeks. Please write comments if you find anything incorrect, or you want to share more information about the topic discussed above.