Academician Liu Jingnan discusses the relationship between natural intelligence and artificial intelligence

Liu Jingnan believes that can understand and define artificial intelligence from the perspective of natural intelligence and time and space.

Currently, information networks are evolving towards the Internet of Things and ubiquitous networks. It is driven by big data, more powerful computing capabilities and more intelligent computing methods, resulting in a new generation of artificial intelligence, that is, computers with the ability to learn and Thinking and other abilities. At the same time, in order to realize the intelligent management and collaborative control of the physical world by computer networks, the cyber-physical system CPS (also known as the ubiquitous network) came into being and was integrated with artificial intelligence to accelerate social production and consumption. From industrialization to automation and intelligence, mankind has entered the era of intelligence. Intelligence is the ability of living creatures in nature to perceive, recognize, and adapt to the environment. Living things have intelligence because they can perceive changes in the outside world and recognize the pros and cons of changes. , after making a decision, adjust itself to achieve advantages and avoid disadvantages. To achieve this goal, it is necessary to find the accurate place, time, orientation, posture, etc., to adjust itself. Therefore, the core of biological intelligence in nature is, Based on the precise time and space location of perceiving and recognizing changes in the external world, we can make choices and behaviors that seek advantages and avoid disadvantages.

Therefore, natural intelligence can be defined as the ability of living things to perceive changes in the outside world, learn and remember to form experiences, rise to cognition, make decisions to adapt to changes in the outside world and their own safety needs, and regulate themselves at the accurate time and place. Or partially change the state of the outside world to achieve the goal of seeking advantages and avoiding disadvantages. Intelligence can solve current problems, while wisdom can solve future and unknown problems. Wisdom requires experience and knowledge formed through perception and cognition. Reasoning, predicting changes in the external world at a certain time and place in the future, and performing self-regulation in advance to change the state of oneself or part of the external world in order to seek benefits and avoid harm. This kind of regulation is often based on precise time and space positions. , accurately coordinate multiple behaviors and actions to change the outside world. Therefore, the ability to position, navigate, and know the time is the innate survival ability of living things to achieve advantages and avoid disadvantages, which belongs to natural intelligence. In this sense, surveying and mapping is The process and means of perceiving, recording and expressing events that occur at a certain time and place, and then assisting cognition, communication and decision-making, are an extension of the technical methods of intelligent behaviors such as human positioning, navigation and timing. In summary, intelligence and wisdom are abilities that only exist in the biological world.

Artificial intelligence is to impart biological intelligence in nature (including human intelligence and wisdom) to machines and environments through technology and methods, so that machines and the environment can interact with each other. The environment can sense and recognize changes in the outside world, and manage and regulate accordingly to seek advantages and avoid disadvantages.

Defining artificial intelligence on the basis of natural intelligence is more accurate and more universal than the current definition of artificial intelligence used by the computer industry such as the Turing test to "make machines perceive and think like humans". It reflects Natural intelligence includes the connotation of 6 types of intelligence such as perception, learning, cognition, decision-making, regulation and even emotion. Artificial intelligence should include at least 5 types of intelligence connotation in addition to emotion. In fact, perceptual intelligence and swarm intelligence in the animal kingdom are in many It far exceeds that of humans. Giving the perceptual intelligence and group intelligence of the animal kingdom to machines and the environment is a major research direction of artificial intelligence

.

According to CSDN author’s definition:

Natural Intelligence (NI) refers to people producing valuable behaviors through brain calculations and decision-making. These behaviors include human brain thinking and decision-making, ears hearing and judgment, eye vision and judgment, nose smell and judgment, skin touch and judgment, etc., and are reflected in all aspects of human behavior.

Artificial Intelligence (AI) uses machines to replace people and realize intelligent behaviors possessed by people. The sentence can be rewritten like this: It mainly refers to computers, data and related software, and can even cover related intelligent terminal equipment. At present, relatively mature technical directions for artificial intelligence applications include machine gaming (intelligent robots), voice recognition, image recognition (text, fingerprints, faces, etc.), and data analysis and prediction provided by sensors. The main disciplines of artificial intelligence research cover computer science, information theory, cybernetics, automation, bionics, biology, psychology, mathematical logic, linguistics, medicine and philosophy, etc.

Machine Learning (ML) is the scientific study of algorithms and statistical models, which are used by computer systems to effectively perform specific tasks without using explicit instructions, but relying on patterns and reasoning. It is considered a subset of artificial intelligence and is the core of artificial intelligence. Machine learning must "learn" with the help of data. Machine learning can be divided into supervised learning, semi-supervised learning, unsupervised learning and reinforcement learning according to the form.

Deep Learning (DL), (also known as deep structured learning or hierarchical learning) is part of a family of machine learning methods based on learning data representations, rather than task-specific algorithms. Deep learning is inspired by information processing and communication patterns in biological nervous systems, but differs from the structure and function of biological brains. Currently, deep learning architectures, such as deep neural networks, deep belief networks, and recurrent neural networks, have been used in computer vision, speech recognition, natural language processing, audio recognition, social network filtering, machine translation, bioinformatics, drug design, and medicine Image analysis and other fields.

(The above concepts and their relationships are original creations of CSDN blogger "Simple Little Bitter Melon", original link: https://blog.csdn.net/weixin_44482877/article/details/122273597)

The above article selection is a way for the author to explore silicon-based organisms and carbon-based life as a whole, or to explore the cognitive methodology of brain-computer interface, so as to understand bionics and brain neural networks.

The above is the detailed content of Academician Liu Jingnan discusses the relationship between natural intelligence and artificial intelligence. For more information, please follow other related articles on the PHP Chinese website!