Google AlphaGO Challenge - the boundaries of artificial intelligence in humans

Recently, the Google AlphaGO and Human Challenge has become a hot topic, which is the challenge between artificial intelligence and Korean Go player Li Shishi. In fact, from the emergence of artificial intelligence, everything about artificial intelligence has always been a hot topic, such as robots. If robots do not have the key factor of artificial intelligence, it is only an automatic production factor, just a "machine." But because the robot has added the "brain" element of artificial intelligence, it has decided that it has elements of "human" from outside the machine itself, which is very imaginative for humans. The background of the Google AlphaGO and Human Challenge events is not covered in this article, because it is a recent hot spot in the media, whether it is positive, negative, neutral, and various viewpoints. What is important is not the game itself, but the impact of this event on human and artificial intelligence.

Humans imitate the product of God's creative behavior

In the entire Western world audience, the universal acceptance and inheritance is "biblical culture", and one of the most important world views is creationism. That is to say, the source of human beings relies on the creation of God, and the operation of the vast universe comes from the care of God, and the creativity of human beings comes from some of the powers that God has given. From the perspective of this world view, it is very easy to understand the behavior of artificial intelligence in the Western world led by the United States. To put it simply, human beings are imitating the behavior created by God and "creating" a kind of human behavior within the scope of human cognitive ability.

The reason why I define the behavior of artificial intelligence as human-like behavior is because, by its very nature, no matter how powerful artificial intelligence is, it is only a physical level behavior, not a biological level behavior, or Behavior at the level of life sciences. Therefore, the so-called concern of replacing human beings and replacing human beings is only at the physical level, and there is no comparability in the life sciences and biological aspects.

The reason why Google AlphaGO has attracted worldwide attention, in addition to Google has always had a very good market value management ability in the capital market, the breakthrough of its own technology is also the focus of everyone's attention. From the perspective of technical architecture, the biggest feature is to break through the "fixed" program logic of traditional computers. The biggest breakthrough is the integration of self-learning, or deep learning, which means Google AlphaGO and traditional meaning. The computer program on it is very different. Google AlphaGO's technical architecture uses a model that mimics the human brain, and the behavior of this model can be understood as Google relies on its powerful team of scientists to computerize this knowledge within the current human knowledge of brain-related exploration. Knowledge exploration has reached a new height.

Artificial intelligence borders in humans

No matter how the artificial intelligence develops, it is essentially a problem at the physical program level. Even if it has the ability of "self-thinking", the boundary of its thinking is also given and set by the developer. The same is true of the Google AlphaGO product itself. Its precondition is that the developer sets a relatively complex self-learning mode, and then starts various calculations by inputting more than 3,000 kinds of game data. And the key element that determines the "smart" level of Google's AlphaGO product is the developer, not the "intelligence" of Google AlphaGO. In other words, the key factor of the "smart" of Google's AlphaGO products is firstly the self-learning mode set by the developer, followed by the quality of the basic knowledge entered by the developer.

From the perspective of the self-learning mode set by the developer, the result is known to be a conversion from a traditional non-variable program to a variable program, so that the program is no longer limited to a fixed program, and there are many possibilities for change. While traditional computers have long surpassed humans in some computing powers, there is a very important difference between humans and humans in terms of logical thinking and judgment. It can be said that for a long time in the computer, Moore's Law is expanding in the vertical capability dimension of computing power, and Google AlphaGO transcends the boundary of traditional computer computing power, referring to the human neural network model. Design, hope to build the ability of thinking of humanoids. But its boundaries depend on human exploration of the brain's neural network model and the developer's perception and thinking about these issues.

From the level of knowledge of self-learning, the key is also dependent on the quality of the big data that developers and humans themselves build. According to this Google AlphaGO, although the developers entered 3,000 kinds of game styles, the quality of these data determines the basis of their subsequent self-learning. If the input game itself is not a master level, then how to self-learn and self In the game, the result can only be said to be the optimal level in the secondary level.

Google AlphaGO may not win

So, back to this artificial intelligence against human intelligence incidents, the key factors that determine whether Google AlphaGO can win are two: First, the "perfectness" of the self-learning architecture established by developers, but it is not currently on this issue. Optimism, it can be said that it has just made a breakthrough. It takes some time to achieve continuous stability. Second, the source of the basic data. Although it has won the European championship before, the level of Go champions in Europe is still relatively obvious in Asia. The difference, if the basic data of Google AlphaGO has not been effectively upgraded before the game, or in the process of self-learning, there is no difficulty in playing and upgrading with more advanced players. There are still some difficulties in winning.

For Korean player Li Shishi, the biggest challenge is not the level of chess, but the psychological and physiological aspects. In terms of computational analysis capabilities, it can be said that Google AlphaGO has no challenging advantages, but in addition to computing power in the process of Go, humanity's spirituality and intuition are at least at this stage, artificial intelligence is still unable to recognize and understand. of. But computers don't have mood swings, they won't be disturbed by the outside world, and they won't feel tired. But humans are different. We have physical and emotional problems. Therefore, it seems to me that the key to Li Shishi's success in this competition lies in the management of his own psychological and physiological aspects. Therefore, unlike the game between the computer game and the human race, the game with humans can be played through some psychological tactics, but these methods are basically ineffective.

If Li Shishi can manage his own psychological and physiological problems, then I think the possibility of winning this game is very big, because at this stage, Google AlphaGO is just an experimental product explored by artificial intelligence in a new direction, just as Like the previous Google glasses, just bring a product from a lab into the real world for testing. The product itself is not "perfect".

If artificial intelligence wins

Regardless of whether Google AlphaGO can win this game, it will be the biggest historical turning point for the development of artificial intelligence. For the big data world where human beings are currently located, and the upcoming Internet of Things era, we can't cope with the data black hole world with our own computer and recognition capabilities. Human beings urgently need artificial intelligence to understand the user's psychology to help process information. Therefore, from the perspective of our time, I prefer to see Google AlphaGO win the game and make this happen. The lab's products have more experimental space and scenes, and as Google Glass continues to explore, it can eventually help humans expand their capabilities.

Especially in the face of the upcoming smart wear era, when all things, including people, are digitized, and data flows and communication are realized by means of data, then for us humans, this huge data processing itself It has surpassed the computing power of our brains. Therefore, from a certain perspective, artificial intelligence is only a product of the development of the intelligent era, but its name is called artificial intelligence. In other words, artificial intelligence does not develop well and assumes the role of a human assistant, and other similar alternative technologies are bound to emerge. Therefore, from the current point of view, the development, maturity and application of artificial intelligence become more and more important.

But from the reality, this still takes a long way to go. Especially for the current concerns about artificial intelligence, it seems too early to me. At present, the most important thing is to accelerate the development of artificial intelligence. As soon as possible, let artificial intelligence come out from the current level of no artificial intelligence, and enter the intelligent level as soon as possible to help users deal with huge data information.

Solar Lawn Light

Solar Lawn Lamps,Best Solar Lawn Lights,Solar Powered Lawn Lights,Outdoor Solar Lawn Lights

Jiangmen Liangtu Photoelectric Technology Co., Ltd. , https://www.liangtulight.com