Understanding Intelligence With A World Self Model
While there are significant differences between artificial and human intelligence, and the nature of intelligence remains unknown, artificial intelligence has achieved considerable success in a variety of activities. Yutao Yue of China's Institute of Deep Perception Technology compares human and artificial intelligence and proposes that a specific aspect of human intelligence is the key to connecting perception and cognition, and that the lack of a new model is impeding understanding and next-level implementation of intelligence. He offered the larger definition of "concept," the concepts and mathematical foundations of the new model of the World-Self Model of intelligence, and eventually a unified universal intelligence framework based on the World-Self Model. The new World-Self Model of Intelligence gives us a better idea of how intelligence works, no matter what kind of job we do or how we look into it.
The ability to generate an informational representation of the physical environment is unique to humans. When we look at an apple, the refracted light passes through the cornea, is controlled by structures like the pupil and crystalline lens, and then translated into electric impulses by retina cells. The apple's physical existence is now represented by a data cluster of electric impulses. The human eye has a resolution of 200 million pixels, although the number of optical nerve cells is just 1 million. Before reaching the visual parts of the brain, the information is compressed over 100 times. Multiple brain visual areas process the compressed representation, which results in higher-level properties such as colour pattern, circular curved form, glossiness, size, and so on. The state and connection patterns of particular brain neuron cells are now the informational representation of the physical reality of the apple. The World-Self Model and its mathematical foundation describe how human intelligence systems function and can theoretically be implemented on a computer.
The framework has two main entities: the intelligent system and the world (environment). There are two aspects to an intelligent system. The End-to-End Model is the first component. The End-to-End Model, as the name implies, takes in data and processes it from beginning to end. It correlates to the human intelligent system's Aspect 1 intelligence (responsive system, System 1) or connectionism Artificial Intelligence. The Global-Self Model is the second portion, which is an idea network that includes world conceptions (dark green region) and the unique notion of self. It relates to the human intelligent system's Aspect 2 intelligence (analytical system, System 2). Aspect 3 intelligence, also known as Symbolic Artificial Intelligence, connects the End-to-End Model with the World-Self Model.
In the World-Self Model's concept of "self," there is a dimension that defines the intelligent system's goal, such as a psychological or social objective, such as self-actualization as a teacher educating kids to be good people, or a predefined objective, such as keeping an elderly lady safe for a household robot (in the future). This aspect of the notion "self" implies that it is a component of the objective. The second portion of the purpose comes from the End-to-End Model, which is a more direct (and sometimes simpler) objective like sending a dodge signal to an actuator when a dangerous rapidly approaching item is detected, such as a boxer's punch or an autonomous automobile approaching out-of-control.