The control room for Google AlphaGo during the match against South Korean Go player Lee Sedol. A Chinese team of artificial intelligence researchers said that they want to challenge the AI. (Photo : Twitter)
A team of Chinese artificial intelligence researchers wants to challenge Google's AlphaGo to a match of the classic board game.
According to the scientists from the Chinese Go Team, they plan to issue the challenge by the end of this year. The team made the announcement at the "Understanding the AlphaGo: War Between Man and Machine and Chinese Artificial Intelligence" forum, Yahoo News reported.
However, the team did not elaborate as to what kind of challenge they would make.
Google's AlphaGo, which was developed by the search engine giant's DeepMind, made headlines earlier this year after it beat South Korean Go player Lee Sedol. The 33-year-old Lee is considered as one of the world's top players in the game.
AlphaGo streaked past Lee 4-1 in a series of games organized by Google to demonstrate the AI's capabilities. The machine's win marked the first time an artificial intelligence program beat a human professional player in the game. Many players previously believed that it would take another decade before the feat would happen.
Go, which originated in China some 2,500 years ago, is considered as one of the world's most complex board games. The game involves two opposing players moving black and white pieces on a 19 by 19 grid, with the aim of capturing are many squares as possible.
Experts said that the biggest challenge for AI programs when playing the game is the sheer number of combinations of the pieces. Each of the pieces in the game can have 361 possible locations on the board, resulting in an immense number of configurations, far more than even chess.
AlphaGo dealt with the problem by using two neural networks to reduce the complexity of the game. A neural network is a system of computers that model the human nervous system. When making its move, the AI first selects a handful of potential moves and then simulates them a few dozen of times to determine which one gives the best outcome.
While Google has yet to comment on the challenge, Google CEO Sundar Pichai reportedly arrived in China to visit one of the country's Go training schools. A spokesman for the company said that Pichai was there to better understand the game and the country, Daily Mail reported.
