Actionism
Behaviorism, also known as evolutionism or cybernetics school, is an intelligent simulation method of behavior based on "perception-action". Behaviorism originated from a school of psychology in the 20th century, which believed that behavior is an organism used to adapt to environmental changes. The combination of various body reactions, its theoretical goal is to anticipate and control behavior.
Its core focus is on cybernetics and believes that artificial intelligence comes from cybernetics. Cybernetics thinking has become an important part of the ideological trend of the times as early as the 1940s and 1950s, affecting early artificial intelligence workers. The cybernetics and self-organizing systems proposed by Wiener and McCulloch, and the engineering cybernetics and biological cybernetics proposed by Qian Xuesen and others have influenced many fields. Cybernetics connects the working principles of the nervous system with information theory, control theory, logic, and computers. Early research work focused on simulating the intelligent behavior and role of humans in the control process, such as research on cybernetic systems such as self-optimization, self-adaptation, self-stabilization, self-organization, and self-learning, and carried out "cybernetic animals" Development. By the 1960s and 1970s, certain progress had been made in the research of these cybernetic systems. The seeds of intelligent control and intelligent robots were planted, and intelligent control and intelligent robot systems were born in the 1980s. Behaviorism only appeared in the face of a new school of artificial intelligence at the end of the 20th century, and aroused the interest of many people. The representative author of this school first recommends Brooks' six-legged walking robot, which is regarded as a new generation of "cybernetic animals", a control system based on perception-action mode that simulates insect behavior.
Connectionism
Connectionism, also known as the Bionic School or the School of Physiology, is an intelligent simulation method based on the neural network and the connection mechanism between the networks and the learning algorithm. Its principle is mainly the connection mechanism and learning algorithm between the neural network and the neural network. It is believed that artificial intelligence originated from bionics, especially the study of human brain models.
Its representative result is the brain model created by physiologist McCulloch and mathematical logician Pitts in 1943, namely MP model, which created a new way to imitate the structure and function of the human brain with electronic devices . It starts with neurons and then studies neural network models and brain models, opening up another development path for artificial intelligence. In the 1960s and 1970s, connectionism, especially the study of brain models represented by perceptrons, had an upsurge. Due to the limitations of theoretical models, biological prototypes and technical conditions at the time, brain model research was in the 20th century. From the late 1970s to the early 1980s, it fell into a low ebb. It was not until Professor Hopfield published two important papers in 1982 and 1984 that he proposed to use hardware to simulate neural networks that connectionism regained its head. In 1986, Rumelhart and others proposed a back propagation (BP) algorithm in a multilayer network. Since then, the momentum of connectionism has been vigorous. From model to algorithm, from theoretical analysis to engineering realization, Wei Neural Network Computer goes to market to lay the foundation. At present, the research enthusiasm for artificial neural networks (ANN) is still high, but the research results are not as good as expected.
Symbolism (Symbolism)
Symbolism is an intelligent simulation method based on logical reasoning, also known as logicism, psychology or computer school. Its principles are mainly the hypothesis of physical symbol system and the principle of limited rationality. It has long been in the leading position in artificial intelligence.
Its core focuses on mathematical logic. Mathematical logic developed rapidly from the end of the 19th century, and began to be used to describe intelligent behavior in the 1930s. After the computer appeared, a logical deduction system was implemented on the computer. Its representative result is the heuristic program LT logic theorist, which proved 38 mathematical theorems and showed that computers can be used to study human thought processes and simulate human intelligence activities. It was these symbolists who first adopted the term "artificial intelligence" as early as 1956. Later he developed heuristic algorithm>expert system>knowledge engineering theory and technology, and made great progress in the 1980s. Symbolism has been thriving for a long time, and has made important contributions to the development of artificial intelligence, especially the successful development and application of expert systems. It is of special significance for artificial intelligence to move toward engineering applications and realize the integration of theory with practice. After the emergence of other schools of artificial intelligence, symbolism is still the mainstream faction of artificial intelligence. Representatives of this school include Newell, Simon, and Nilsson.
PS: At present, both symbolism (traditional machine learning) and connectionism (deep learning) have been shining, and the foreseeable behaviorist school will gradually usher in a prosperous period. The reason for this inference is that cybernetics is inseparable from the system, and the simulation of a system is inseparable from a huge amount of computation. When the hardware computing power is further improved, and the simulation operation of some complex systems can be quickly implemented, then phenomena such as the evolution and emergence in the complex system will bring longer progress to artificial intelligence.