News classification
Contact us
- Add: No. 9, North Fourth Ring Road, Haidian District, Beijing. It mainly includes face recognition, living detection, ID card recognition, bank card recognition, business card recognition, license plate recognition, OCR recognition, and intelligent recognition technology.
- Tel: 13146317170 廖经理
- Fax:
- Email: 398017534@qq.com
How much is AI IQ?
How much is AI IQ?
1. Whether artificial intelligence can transcend the wisdom of human beings
And in stark contrast, in October 26th, Softbank CEO Sun Zhengyi held in Saudi Arabia's future investment plan conference that AI's intelligence quotient (IQ) will reach 10000 in the next 30 years. The general assembly robot citizen Sophia (Sophia) was announced as the Saudi machine citizens, these events to discuss the relationship between artificial intelligence and human become more lively, more people worried that robots will eventually make humans into useless creatures, even lead to the extinction of human.
AI
It includes famous physicists Hocking, Microsoft founder Bill Gates, Tesla CEO musk, and the founder of the Softbank group Sun Zhengyi. The point of view is that artificial intelligence will surpass human beings and pose a threat to human beings.
Faced with the sudden outbreak of AI, widely spread in artificial intelligence threat, so AI really can not go beyond the human intelligence, AI level to what level, what kinds of intelligent products have many intelligent, explosive AI and robot breakthrough event whether true?
These problems need to be set up scientifically in a scientific way, which can make a unified intelligence level test for AI system, human beings and other agents, so as to judge their level of development.
At present, the quantitative evaluation of artificial intelligence faces two important challenges: first, the AI system has not yet formed a unified model; second, there is no unified model between AI system and human beings represented by life.
The two challenge all point to the same problem, namely the system of artificial intelligence all and all life (especially to human life as the representative) need a unified model to describe, this is the only way in this model is established on intelligence measuring method for testing.
Since the Turing test was put forward in 1950, scientists have done a lot of work for the evaluation system of artificial intelligence development.
In 1950, Turing presented a famous Turing experiment, using questions and human referees to determine whether a computer has the same human intelligence. As artificial intelligence test method is the most widely used, but the Turing test does not test the intelligence development level of Ai, the intelligent system can determine just the same with human intelligence, and artificial interference too much, depends heavily on the subjective judgment of the referee and the testers, so people often have not been rigorously validated in the case under the program that pass the Turing test,
In March 24, 2015, the Journal of the American Academy of Sciences (PNAS) published a paper, and put forward a new Turing Test Method "Visual Turing test", which is used to further evaluate the image recognition ability of computers.
In 2014, Professor Reed (Mark O. Riedl) of the Georgia Institute of technology of the United States believed that the essence of intelligence was creativity. He designed a test called Lovelace 2. The test range of Lovelace 2 includes the creation of fictions with fictitious stories, poetic creation, oil painting and music.
There are two problems in solving the problem of quantitative test of artificial intelligence, including Turing test, including: first, these tests do not form a unified intelligent model, and based on it, we analyze and distinguish multiple classifications of intelligence. Will lead to different intelligent systems including human unified testing; second is the test method to quantitative analysis of artificial intelligence, or only a quantitative analysis of intelligence, but the system to achieve what percentage of human wisdom, how to develop speed and wisdom development speed ratio, these problems not involved in in the study.
Three. Establishment of standard intelligent model and AI IQ test
To solve this problem, Liu Feng, Shi Yong and Liu Ying team of the virtual economy and data science research center of the Chinese Academy of Sciences refer to Feng Neumann structure, David Wechsler's human intelligence model and DIKW model system in knowledge management field. Since 2014, a "standard intelligent model" has been set up to describe the characteristics and attributes of artificial intelligence systems and human beings.
The von Neumann structure? Give us the inspiration is: the standard model for intelligent systems should include input and output system, to get information from the outside world, can be generated inside the feedback to the outside world, only in this way, the standard of intelligent system can become a "live" system.
Artificial intelligence
David Wexler on human intelligence is defined to give us the biggest revelation is: intellectual ability is composed of multiple elements, rather than the Turing test or visual Turing test that focus on one aspect of intellectual ability.
The inspiration from DIKW model system is that intelligence is a ability to solve problems and accumulate knowledge. Knowledge is the structured data and information precipitated by human interaction with the outside world, which reminds us that an intelligent system is not only reflected.
Since twenty-first Century, many important events have been broken out in the field of artificial intelligence. Among the most attractive, is the AlphaGo that defeated the human Weiqi champion in 2016 and began to learn from 0.
And in stark contrast, in October 26th, Softbank CEO Sun Zhengyi held in Saudi Arabia's future investment plan conference that AI's intelligence quotient (IQ) will reach 10000 in the next 30 years. The general assembly robot citizen Sophia (Sophia) was announced as the Saudi machine citizens, these events to discuss the relationship between artificial intelligence and human become more lively, more people worried that robots will eventually make humans into useless creatures, even lead to the extinction of human.
AI
It includes famous physicists Hocking, Microsoft founder Bill Gates, Tesla CEO musk, and the founder of the Softbank group Sun Zhengyi. The point of view is that artificial intelligence will surpass human beings and pose a threat to human beings.
On the other hand, scientists in the field of artificial intelligence have also put forward refutation or disagreement on the threat theory of artificial intelligence. This includes YannLeCun, director of the Facebook Artificial Intelligence Laboratory, and Wu Enda, the former founder of the Google brain.
Faced with the sudden outbreak of AI, widely spread in artificial intelligence threat, so AI really can not go beyond the human intelligence, AI level to what level, what kinds of intelligent products have many intelligent, explosive AI and robot breakthrough event whether true?
These problems need to be set up scientifically in a scientific way, which can make a unified intelligence level test for AI system, human beings and other agents, so as to judge their level of development.
At present, the quantitative evaluation of artificial intelligence faces two important challenges: first, the AI system has not yet formed a unified model; second, there is no unified model between AI system and human beings represented by life.
The two challenge all point to the same problem, namely the system of artificial intelligence all and all life (especially to human life as the representative) need a unified model to describe, this is the only way in this model is established on intelligence measuring method for testing.
Since the Turing test was put forward in 1950, scientists have done a lot of work for the evaluation system of artificial intelligence development.
In 1950, Turing presented a famous Turing experiment, using questions and human referees to determine whether a computer has the same human intelligence. As artificial intelligence test method is the most widely used, but the Turing test does not test the intelligence development level of Ai, the intelligent system can determine just the same with human intelligence, and artificial interference too much, depends heavily on the subjective judgment of the referee and the testers, so people often have not been rigorously validated in the case under the program that pass the Turing test,
In March 24, 2015, the Journal of the American Academy of Sciences (PNAS) published a paper, and put forward a new Turing Test Method "Visual Turing test", which is used to further evaluate the image recognition ability of computers.
In 2014, Professor Reed (Mark O. Riedl) of the Georgia Institute of technology of the United States believed that the essence of intelligence was creativity. He designed a test called Lovelace 2. The test range of Lovelace 2 includes the creation of fictions with fictitious stories, poetic creation, oil painting and music.
There are two problems in solving the problem of quantitative test of artificial intelligence, including Turing test, including: first, these tests do not form a unified intelligent model, and based on it, we analyze and distinguish multiple classifications of intelligence. Will lead to different intelligent systems including human unified testing; second is the test method to quantitative analysis of artificial intelligence, or only a quantitative analysis of intelligence, but the system to achieve what percentage of human wisdom, how to develop speed and wisdom development speed ratio, these problems not involved in in the study.
Three. Establishment of standard intelligent model and AI IQ test
To solve this problem, Liu Feng, Shi Yong and Liu Ying team of the virtual economy and data science research center of the Chinese Academy of Sciences refer to Feng Neumann structure, David Wechsler's human intelligence model and DIKW model system in knowledge management field. Since 2014, a "standard intelligent model" has been set up to describe the characteristics and attributes of artificial intelligence systems and human beings.
The von Neumann structure? Give us the inspiration is: the standard model for intelligent systems should include input and output system, to get information from the outside world, can be generated inside the feedback to the outside world, only in this way, the standard of intelligent system can become a "live" system.
Artificial intelligence
David Wexler on human intelligence is defined to give us the biggest revelation is: intellectual ability is composed of multiple elements, rather than the Turing test or visual Turing test that focus on one aspect of intellectual ability.
The inspiration from DIKW model system is that intelligence is a ability to solve problems and accumulate knowledge. Knowledge is the structured data and information precipitated by human interaction with the outside world, which reminds us that an intelligent system is not only reflected.