A photo from open sources
Robots created for logic games with humans are not too impress ordinary layman. However, do not rush with conclusions. The fact is that such programs demonstrate today’s most impressive self-study results, self-improvement and imitation of human thinking.
For example, artificial intelligence “AlphaGo”, which will be discussed speech in this news, became famous in May of this year for the fact that the first to beat the world professional in Go – the popular desktop a game with the deepest strategic content. And he did it without any giveaway on the part of man. This victory has become a very significant breakthrough in the field of AI, since scientists, well versed in Go, predicted that it will happen no earlier 2025 year.
Formerly, AlphaGo training used the general principles. machine learning and elementary game theories in Go, however with recently, the robot has refused any help from its creators and started developing completely independently.
A photo from open sources
Someone will be surprised: what is so outstanding about this, and how is it at all this program can be called advanced artificial intelligence? Like, iterates through the algorithm many options the arrangement of stones on the board and finds optimal solutions for victory. Business … However, in fact, everything is much more complicated and more interesting.
“AlphaGo” operates on the basis of neural networks, that is, it imitates by and large, the mental activity of man. In fact in fact, the program does not iterate over any options, because their number Go is so great for the game that it can’t cope with this task millions of years, even the most powerful modern supercomputer. Artificial Intelligence uses logic and a kind of semblance intuition, he also bluffs and defames in a purely human way. Playing with by people, he constantly studies human behavior, borrowing from living players smart moves, remembering unsuccessful, perhaps absorbing even human emotions.
The program has already learned to invent its own unique game methods that allow her to easily beat the world champions. Recognized craftsmen have to hone their strategy many years to develop similar steps. AlphaGo has it takes a few days. Unlike other programs, designed for logic games, AlphaGo not only uses human knowledge, but also significantly beyond their scope. Some seasoned Go players admit that the program plays on level of God, “and competing with her is no longer possible to win, and to learn from her skill. Like this, dear skeptics who do not believe that artificial intelligence is already soon equal to human, and then surpass it!
Artificial Intelligence Robots
