Product Code Database
In the philosophy of artificial intelligence, GOFAI ( good old-fashioned artificial intelligence) is classical symbolic AI, as opposed to other approaches, such as neural networks, situated robotics, narrow AI symbolic AI or neuro-symbolic AI. The term was coined by philosopher John Haugeland in his 1985 book Artificial Intelligence: The Very Idea.
Haugeland coined the term to address two questions:
AI research in the 1950s and 60s had an enormous influence on intellectual history: it inspired the cognitive revolution, led to the founding of the academic field of cognitive science, and was the essential example in the philosophical theories of computationalism, functionalism and cognitivism in ethics and the psychological theories of cognitivism and cognitive psychology. The specific aspect of AI research that led to this revolution was what Haugeland called "GOFAI".
In AI development and technology, GOFAI is used to refer to programs that are built with deliberate, explicit instructions for a single task. This is in contrast to approaches that use machine learning. Examples of GOFAI applications include AlphaGo and Apple's initial Siri design.
Symbolic AI in the 1960s was able to successfully simulate the process of high-level reasoning, including logical deduction, algebra, geometry, spatial reasoning and means-ends analysis, all of them in precise English sentences, just like the ones humans used when they reasoned. Many observers, including philosophers, psychologists and the AI researchers themselves became convinced that they had captured the essential features of intelligence. This was not just hubris or speculation -- this was entailed by rationalism. If it was not true, then it brings into question a large part of the entire Western philosophical tradition.
Continental philosophy, which included Nietzsche, Edmund Husserl, Martin Heidegger and others, rejected rationalism and argued that our high-level reasoning was limited and prone to error, and that most of our abilities come from our intuitions, culture, and instinctive feel for the situation. Philosophers who were familiar with this tradition were the first to criticize GOFAI and the assertion that it was sufficient for intelligence, such as Hubert Dreyfus and Haugeland.
Haugeland coined the term GOFAI in order to examine the philosophical implications of “the claims essential to all GOFAI theories”, which he listed as:
This is very similar to the sufficient side of the physical symbol systems hypothesis proposed by Herbert A. Simon and Allen Newell in 1963:
It is also similar to Hubert Dreyfus' "psychological assumption":
Haugeland's description of GOFAI refers to symbol manipulation governed by a set of instructions for manipulating the symbols. The "symbols" he refers to are discrete physical things that are assigned a definite semantics -- like <cat> and <mat>. They do not refer to signals, or unidentified numbers, or matrixes of unidentified numbers, or the zeros and ones of digital machinery. Thus, Haugeland's GOFAI does not include "good old fashioned" techniques such as , perceptrons, dynamic programming or control theory or modern techniques such as neural networks or support vector machines.
These questions ask if GOFAI is sufficient for general intelligence -- they ask if there is nothing else required to create fully intelligent machines. Thus GOFAI, for Haugeland, does not include systems that combine symbolic AI with other techniques, such as neuro-symbolic AI, and also does not include narrow AI symbolic AI systems that are designed only to solve a specific problem and are not expected to exhibit general intelligence.
The technology they criticized came to be called Good Old-Fashioned AI (GOFAI). GOFAI corresponds to the simplest logical agent design ... and we saw ... that it is indeed difficult to capture every contingency of appropriate behavior in a set of necessary and sufficient logical rules; we called that the qualification problem.
Later symbolic AI work after the 1980's incorporated more robust approaches to open-ended domains such as probabilistic reasoning, non-monotonic reasoning, and machine learning.
Currently, most AI researchers believe deep learning, and more likely, a synthesis of neural and symbolic approaches (neuro-symbolic AI), will be required for general intelligence.
|
|
