Google DeepMind says it has influenced another huge progress in artificial intelligence by getting a machine to ace the Chinese game of Go without any human players to offer assistance. The tech mammoth's AI division conceived the AlphaGo program, and had it beat two of the world's best players. It began by learning from many games played by people, however the new AlphaGo started with a clear Go board and no data other than the rules, and after that played itself. It is reported, that inside of 72 hours it was able to beat the first program by 100 games to zero. Demis Hassabis, CEO of DeepMind said, the framework could now have more broad applications in scientific research. He told BBC," even though we are obviously a long way from full AI, we think this is now good enough to make some real progress on some real problems."
Also, it is reported that the London-based artificial intelligence organization's product defeated leading South Korean Go player Lee Se-dol by four games to one last year.
Where the game has more conceivable moves than are atoms in the universe, this was a major triumph for machine over man, coming sooner than individuals in the AI world anticipated. David Silver, who was the driving force behind the program, says the group adopted an altogether different strategy with AlphaGo Zero. He clarified, “the new version starts from a neural network that knows nothing at all about the game of Go.”
Go was thought to go back to feudal China thousands of years ago. Go utilizes white and dark stones on a gird, similar to chess, only in that instead of battling for each-others king piece, the players grapple for ‘territory’ on the Go board. Go's principles are less difficult to comprehend compared to chess, yet the quantity of moves in chess is around 20, contrasted with Go, that has a decision pool of around 200 moves. It can be extremely hard to figure out who is winning in Go, and this has ended up being a more productive method for displaying the capabilities of modern AI.
The AIphaGo Zero created methods that turned out to be so intriguing and powerful that a significant number of the group has proceeded onward to new ventures, attempting to take this technology to new heights. Demis Hassabis specified medication design and the discovery of new materials as regions of interest.
