In conclusion, let’s discuss the approaches that researchers pursued using electronic digital programmable computers.

From the 1960s through the 1970s, symbolic approaches achieved success at simulating high-level thinking in specific application programs. For example, in 1963, Danny Bobrow’s technical report from MIT’s AI group proved that computers could understand natural language well enough to solve algebra word problems correctly. The success of symbolic approaches added credence to the belief that symbolic approaches eventually would succeed in creating a machine with artificial general intelligence, also known as “strong AI,” equivalent to a human mind’s intelligence.

By the 1980s, however, symbolic approaches had run their course and fallen short of the goal of artificial general intelligence. Many AI researchers felt symbolic approaches never would emulate the processes of human cognition, such as perception, learning, and pattern recognition. The next step was a small retreat, and a new era of AI research termed “subsymbolic” emerged. Instead of attempting general AI, researchers turned their attention to solving smaller specific problems. For example researchers such as Australian computer scientist and former MIT Panasonic Professor of Robotics Rodney Brooks rejected symbolic AI. Instead he focused on solving engineering problems related to enabling robots to move.

In the 1990s, concurrent with subsymbolic approaches, AI researchers began to incorporate statistical approaches, again addressing specific problems. Statistical methodologies involve advanced mathematics and are truly scientific in that they are both measurable and verifiable. Statistical approaches proved to be a highly successful AI methodology. The advanced mathematics that underpin statistical AI enabled collaboration with more established fields, including mathematics, economics, and operations research. Computer scientists Stuart Russell and Peter Norvig describe this movement as the victory of the “neats” over the “scruffies,” two major opposing schools of AI research. Neats assert that AI solutions should be elegant, clear, and provable. Scruffies, on the other hand, assert that intelligence is too complicated to adhere to neat methodology.

From the 1990s to the present, despite the arguments between neats, scruffies, and other AI schools, some of AI’s greatest successes have been the result of combining approaches, which has resulted in what is known as the “intelligent agent.” The intelligent agent is a system that interacts with its environment and takes calculated actions (i.e., based on their success probability) to achieve its goal. The intelligent agent can be a simple system, such as a thermostat, or a complex system, similar conceptually to a human being. Intelligent agents also can be combined to form multiagent systems, similar conceptually to a large corporation, with a hierarchical control system to bridge lower-level subsymbolic AI systems to higher-level symbolic AI systems.

The intelligent-agent approach, including integration of intelligent agents to form a hierarchy of multiagents, places no restriction on the AI methodology employed to achieve the goal. Rather than arguing philosophy, the emphasis is on achieving results. The key to achieving the greatest results has proven to be integrating approaches, much like a symphonic orchestra integrates a variety of instruments to perform a symphony.

In the last seventy years, the approach to achieving AI has been more like that of a machine gun firing broadly in the direction of the target than a well-aimed rifle shot. In fits of starts and stops, numerous schools of AI research have pushed the technology forward. Starting with the loftiest goals of emulating a human mind, retreating to solving specific well-defined problems, and now again aiming toward artificial general intelligence, AI research is a near-perfect example of all human technology development, exemplifying trial-and-error learning, interrupted with spurts of genius.

Although AI has come a long way in the last seventy years and has been able to equal and exceed human intelligence in specific areas, such as playing chess, it still falls short of general human intelligence or strong AI. There are two significant problems associated with strong AI. First, we need a machine with processing power equal to that of a human brain. Second, we need programs that allow such a machine to emulate a human brain.

Source: The Artificial Intelligence Revolution (2014), Louis A. Del Monte

In the last post (Part 1/3), we made the point that no unifying theory guides AI research. Researchers disagree among themselves, and we have more questions than answers. Here are two major questions that still haunt AI research.

1. Should AI simulate human intelligence, incorporating the sciences of psychology and neurology, or is human biology irrelevant?
2. Can AI, simulating a human mind, be developed using simple principles, such as logic and mechanical reasoning, or does it require solving a large number of completely unrelated problems?

Why do the above questions still haunt AI? Let us take some examples.

• Similar types of questions arose in other scientific fields. For example, in the early stages of aeronautics, engineers questioned whether flying machines should incorporate bird biology. Eventually bird biology proved to be a dead end and irrelevant to aeronautics.
• When it comes to solving problems, humans rely heavily on our experience, and we augment it with reasoning. In business, for example, for every problem encountered, there are numerous solutions. The solution chosen is biased by the paradigms of those involved. If, for example, the problem is related to increasing the production of a product being manufactured, some managers may add more people to the work force, some may work at improving efficiency, and some may do both. I have long held the belief that for every problem we face in industry, there are at least ten solutions, and eight of them, although different, yield equivalent results. However, if you look at the previous example, you may be tempted to believe improving efficiency is a superior (i.e., more elegant) solution as opposed to increasing the work force. Improving efficiency, however, costs time and money. In many cases it is more expedient to increase the work force. My point is that humans approach solving a problem by using their accumulated life experiences, which may not even relate directly to the specific problem, and augment their life experiences with reasoning. Given the way human minds work, it is only natural to ask whether intelligent machines will have to approach problem solving in a similar way, namely by solving numerous unrelated problems as a path to the specific solution required.

Scientific work in AI dates back to the 1940s, long before the AI field had an official name. Early research in the 1940s and 1950s focused on attempting to simulate the human brain by using rudimentary cybernetics (i.e., control systems). Control systems use a two-step approach to controlling their environment.

1. An action by the system generates some change in its environment.
2. The system senses that change (i.e., feedback), which triggers the system to change in response.

A simple example of this type of control system is a thermostat. If you set it for a specific temperature, for example 72 degrees Fahrenheit, and the temperature drops below the set point, the thermostat will turn on the furnace. If the temperature increases above the set point, the thermostat will turn off the furnace. However, during the 1940s and 1950s, the entire area of brain simulation and cybernetics was a concept ahead of its time. While elements of these fields would survive, the approach of brain simulation and cybernetics was largely abandoned as access to computers became available in the mid-1950s.

With access to electronic digital programmable computers in the mid-1950s, AI researchers began to focus on symbol manipulation (i.e., the manipulation of mathematical expressions, as is found in algebra) to emulate human intelligence. Three institutions led the charge: Carnegie Mellon University, Stanford University, and the Massachusetts Institute of Technology (MIT). Each university had its own style of research, which the American philosopher John Haugeland (1945–2010) named “good old-fashioned AI” or “GOFAI.”

In the conclusion to this article (Part 3/3), we will discuss the approaches that researchers pursued using electronic digital programmable computers.

Source: The Artificial Intelligent Revolution (2014), Louis A. Del Monte