Computer Go is the field of artificial intelligence (AI) dedicated to creating a computer program that plays the traditional board gameGo. The field is sharply divided into two eras. Before 2015, the programs of the era were weak. The best efforts of the 1980s and 1990s produced only AIs that could be defeated by beginners, and AIs of the early 2000s were intermediate level at best. Professionals could defeat these programs even given handicaps of 10+ stones in favor of the AI. Many of the algorithms such as alpha-beta minimax that performed well as AIs for checkers and chess fell apart on Go's 19x19 board, as there were too many branching possibilities to consider. Creation of a human professional quality program with the techniques and hardware of the time was out of reach. Some AI researchers speculated that the problem was unsolvable without creation of human-like AI.
The application of Monte Carlo tree search to Go algorithms provided a notable improvement in the late 2000s decade, with programs finally able to achieve a low-dan level: that of an advanced amateur. High-dan amateurs and professionals could still exploit these programs' weaknesses and win consistently, but computer performance had advanced past the intermediate (single-digit kyu) level. The tantalizing unmet goal of defeating the best human players without a handicap, long thought unreachable, brought a burst of renewed interest. The key insight proved to be an application of machine learning and deep learning. DeepMind, a Google acquisition dedicated to AI research, produced AlphaGo in 2015 and announced it to the world in 2016. AlphaGo defeated Lee Sedol, a 9 dan professional, in a no-handicap match in 2016, then defeated Ke Jie in 2017, who at the time continuously held the world No. 1 ranking for two years. Just as checkers had fallen to machines in 1995 and chess in 1997, computer programs finally conquered humanity's greatest Go champions in 2016–2017. DeepMind did not release AlphaGo for public use, but various programs have been built since based on the journal articles DeepMind released describing AlphaGo and its variants. (Full article...)
The following are images from various Go-related articles on Wikipedia.
Image 1The illustration displays the four "liberties" (adjacent empty points) of a single black stone. Illustrations , , and show White reducing those liberties progressively by one. In , when Black has only one liberty left, that stone is under attack and about to be captured and eliminated (a state called atari). White may capture that stone (remove it from the board) with a play on its last liberty (at D-1). (from Go (game))
Image 2The first 150 moves of a Go game animated. (Click on the board to restart the animation in a larger window.) (from Go (game))
Image 3A 9×9 game with graphical aids. Colors and markings show evaluations by the computer assistant. (from Go (game))
Image 4Three Japanese professional Go players observe some younger amateurs as they dissect a life and death problem in the corner of the board, at the US Go Congress in Houston, Texas, 2003. (from Go (game))
Image 5Under normal rules, White cannot play at A because that point has no liberties. Under the Ing and New Zealand rules, White may play A, a suicide stone that kills itself and the two neighboring white stones, leaving an empty three-space eye. Black naturally answers by playing at A, creating two eyes to live. (from Go (game))
Image 8One black chain and two white chains, with their liberties marked with dots. Liberties are shared among all stones of a chain and can be counted. Here the black group has 5 liberties, while the two white chains have 4 liberties each. (from Go (game))
Image 9A traditional Japanese set, with a solid wooden floor board (碁盤goban), 2 bowls (碁笥goke) and 361 stones (碁石goishi) (from Go (game))
Image 10A finished beginner's game on a 13×13 board (from Go (game))
Image 12An example of single-convex stones and Go Seigen bowls. These particular stones are made of Yunzi material, and the bowls of jujube wood. (from Go (game))
Image 13The Black stone group has only one liberty (at point A), so it is very vulnerable to capture. If Black plays at A, the chain would then have 3 liberties, and so is much safer. However, if White plays at A first, the Black chain loses its last liberty, and thus it is captured and immediately removed from the board, leaving White's stones as shown to the right. (from Go (game))
Image 15Go portrayed as part of East-Asian culture. (The goblet in the middle is from the Nihon Ki-in.) (from Go (game))
Image 16A simplified game at its end. Black's territory (A) + (C) and prisoners (D) is counted and compared to White's territory (B) only (no prisoners). In this example, both Black and White attempted to invade and live (C and D groups) to reduce the other's total territory. Only Black's invading group (C) was successful in living, as White's group (D) was killed with a black stone at (E). The points in the middle (F) are dame, meaning they belong to neither player. (from Go (game))
Image 18Painting of a woman playing Go, from the Astana Graves. Tang dynasty, c. 744 CE. (from Go (game))
Image 19Example of seki (mutual life). Neither Black nor White can play on the marked points without reducing their own liberties for those groups to one (self-atari). (from Go (game))
Image 20Go players demonstrating the traditional technique of holding a stone (from Go (game))
Image 21Minamoto no Yoshiie by Tsukioka Yoshitoshi, 1886. This popular woodblock print depicts the ancient legend of a husband who suspected his wife was having an affair with the samurai Minamoto no Yoshiie. To prevent his visits, the husband surrounded his house with brambles and placed a Go board on the balcony, hoping he would stumble over it. Instead, the samurai deftly cut the board as he leaped over the balcony railing, avoiding both obstacles. (from Go (game))
Image 22Model of a 19×19 Go board, from a tomb of the Sui dynasty (581–618 CE) (from Go (game))
Image 23A simplified ko fight on a 9×9 board. The ko is at the point marked with a square—Black has "taken the ko" first. The ko fight determines the life of the A and B groups—only one survives and the other is captured. White may play C as a ko threat, and Black properly answers at D. White can then take the ko by playing at the square-marked point (capturing the one black stone). E is a possible ko threat for Black. (from Go (game))
The dan (段) ranking system is used by many Japanese, Okinawan, Korean, and other martial arts organizations to indicate the level of a person's ability within a given system. Used as a ranking system to quantify skill level in a specific domain, it was originally used at a Go school during the Edo period. It is now also used in most modern Japanese fine and martial arts. (Full article...)
Image 3
AlphaGo is a computer program that plays the board gameGo. It was developed by the London-based DeepMind Technologies, an acquired subsidiary of Google. Subsequent versions of AlphaGo became increasingly powerful, including a version that competed under the name Master. After retiring from competitive play, AlphaGo Master was succeeded by an even more powerful version known as AlphaGo Zero, which was completely self-taught without learning from human games. AlphaGo Zero was then generalized into a program known as AlphaZero, which played additional games, including chess and shogi. AlphaZero has in turn been succeeded by a program known as MuZero which learns without being taught the rules. (Full article...)
The rules of Go have seen some variation over time and from place to place. This article discusses those sets of rules broadly similar to the ones currently in use in East Asia. Even among these, there is a degree of variation. (Full article...)
Image 7
Hikaru no Go (ヒカルの碁, lit. Hikaru's Go) is a Japanese manga series based on the board game Go, written by Yumi Hotta and illustrated by Takeshi Obata. The production of the series' Go games was supervised by Go professional Yukari Umezawa. It was serialized in Shueisha's Weekly Shōnen Jump from 1998 to 2003, with its chapters collected into 23 tankōbon volumes. The story follows Hikaru, who discovers a Go board in his grandfather's attic one day. The object turns out to be haunted by a ghost named Sai, the emperor's former Go teacher in the Heian era. Sai finds himself trapped in Hikaru's mind and tells him which moves to play against opponents, astonishing onlookers with the boy's apparent level of skill at the game. (Full article...)
Image 8
AlphaGo versus Lee Sedol, also known as the DeepMind Challenge Match, was a five-game Go match between top Go player Lee Sedol and AlphaGo, a computer Go program developed by DeepMind, played in Seoul, South Korea between 9 and 15 March 2016. AlphaGo won all but the fourth game; all games were won by resignation. The match has been compared with the historic chess match between Deep Blue and Garry Kasparov in 1997. (Full article...)
There are various systems of Go ranks and ratings that measure the skill in the traditional board game Go. Traditionally, Go rankings have been measured using a system of dan and kyu ranks. Especially in amateur play, these ranks facilitate the handicapping system, with a difference of one rank roughly corresponding to one free move at the beginning of the game. This system is also commonly used in many East Asian martial arts, where it often corresponds with a belt color. With the ready availability of calculators and computers, rating systems have been introduced. In such systems, a rating is rigorously calculated on the basis of game results. (Full article...)
Zdroj:https://en.wikipedia.org?pojem=Portal:Go Text je dostupný za podmienok Creative Commons Attribution/Share-Alike License 3.0 Unported; prípadne za ďalších podmienok. Podrobnejšie informácie nájdete na stránke Podmienky použitia.