On March 15, 2016 (February 7, 2016 lunar calendar), "man-machine war" final battle Lee Sedol lost the total score of 1 to 4 AlphaGo. Google Go artificial intelligence AlphaGo and South Korean chess player Lee Sedol in the afternoon of March 15, 2016 Beijing time, Google Go artificial intelligence AlphaGo played the last round of competition with South Korean chess player Lee Sedol today. AlphaGo won the game, and the final man-machine war total score froze frame at 1:4. The score was regained to 3:1. Lee Sedol said after the game that there were loopholes in AlphaGo's black control, and he would be black in the last game. In the fourth game, Lee Se-dol played Bai 78 "God's hand", which caused AlphaGo to misjudge the situation. It was not until 5 moves later that he realized that the situation was unfavorable. Lee Se-dol won the middle game and won the first victory against AlphaGo. And AlphaGo also ranked fourth in the world because of the first loss. Today, Lee Se-dol entered the game with his wife and daughter, looking quite relaxed. After AlphaGo won 3:0, Lee Se-dol obviously let go of his mental burden. Lee Se-dol is with his wife and daughter. In this game, Lee Se-dol is the black first, and Google AlphaGo is the white one. The first battle between Lee Se-dol and AlphaGo is on the top right and bottom right of the board. Li tenaciously fights to make this part of the black chess live. After that, the two sides fight in the center of the board. When the game was 3 hours and 40 minutes, Lee Sedol's time was used up and the game entered the countdown stage. However, the time difference between the two sides was not large today, and AlphaGo only had 20 minutes left at this time. Unlike the previous 4 games, today's matchup did not result in one side winning in the middle of the game, and the two sides fought until the closing stage. In the end, after 5 hours of the game, Lee Sedol threw in 180 hands. Because AlphaGo defeated Lee Sedol with a total score of 4:1, Google will set up a prize of 1 million US dollars. Google has announced that it will donate this prize to UNICEF and related Go education institutions. AlphaGo's development team, DeepMind, expressed its willingness to come to China after the first game, and may organize a showdown with Ke Jie, who is currently ranked No. 1 in the world. The fifth game of the human-machine war (Lee Sedol's black) AlphaGo's five games are over. Where will artificial intelligence go in the future? After the first human-machine war, Demis Hassabis, the father of AlphaGo and co-founder of DeepMind, said that the next goal of artificial intelligence is to let computers learn chess by themselves. That is to say, the next version of AlphaGo will start from scratch, do not accept the specific knowledge instilled by humans, and achieve true autonomous learning. As for poker and computer games played by artificial intelligence, which are of public concern, Hassabis said that Go is a "complete information game", in which all the information on both sides of the game is presented on the board. Whereas poker and computer games, which are played by multiple players, are "incomplete information games", in which the computer cannot know all the information, and these areas are still difficult for artificial intelligence. DeepMind believes that the ultimate goals include smartphone assistants, health care and robotics. And while AlphaGo is currently just a system for playing Go, Hassabis believes that some basic principles also apply to solving real-world problems. Eric Schmidt, chairperson of Google's parent company Alphabet, said: "I don't think there is any area where this technology is not applicable." He cited Google's traditional search and advertising business, driverless cars and Verily, its health unit. "For me, this technology will be used in every company at Alphabet." In the past week, AlphaGo's victory over top human players has caused people to feel disappointed and worried. Schmidt explained to the public before the game to allay doubts: "No matter what the final result is, the winner is human." Related information AlphaGoAlphaGo is a Go artificial intelligence program developed by Google's DeepMind company. In October 2015, the alpha coefficient Go game defeated the European Go champion and professional second-stage player Fan Hui 5-0. Lee Sedol Lee Sedol entered the stage in 1995, and was directly promoted to the ninth stage after winning the 16th Fujitsu Cup in July 2003. 2006, 2007, 2008 Korean Go Awards - Best Player Award (MVP).