What about Go?

Go, a well-liked board game in China and Japan, poses a special challenge to artificial intelligence. Contrary to chess, current Go programs are unable to compete in the game despite significant efforts. This is because in Go, players must mentally separate a position into numerous sub positions and examine each one separately before taking their interactions into account. Although people still play chess using this strategy, modern chess programs just look at the position as a whole. Chess systems carry out massive computations, often numbering in the millions, as shown in Deep Blue, to make up for this lack of cognitive abilities. However, it won’t be long before AI research fixes this flaw and creates tools that are proficient enough to play Go.

Over 2,500 years have passed since the Chinese first played the board game go. It is thought to have been created during the Zhou dynasty and has since taken root in nations throughout East Asia, including Japan, Korea, and Taiwan. Chess and go are frequently contrasted, although go is a far more difficult game with a bigger board and more pieces. The goal of the game, which is played on a 19×19 grid, is to encircle and capture the other player’s pieces while defending one’s own.

In comparison to the number of atoms in the known cosmos, there are more conceivable board configurations in the game of Go. As a result, both playing and programming the game are incredibly difficult. While chess computer programs have long been capable of outplaying the best human players, the same cannot be said for Go computer programs. In fact, up until recently, even the finest Go programs couldn’t defeat novice players.

Go’s enormous number of possible moves on each turn is one of the reasons it is so challenging for computers to play. Go allows players to place stones anywhere on the board as long as the space is unoccupied, unlike chess where the number of pieces on the board limits the number of movements that may be made. This indicates that there are millions of potential moves for each turn, and a computer is unable to consider them all.

Go is difficult for computers to play because it is a game of strategy and intuition rather than calculation. Players in chess can rely on well-known opening movements and strategies, while there are no predetermined openings or patterns in go. To make the best decisions, players must rely on their instinct and strategic reasoning, which is something that computers struggle with.

However, recent developments in AI have significantly improved Go-playing software. In a five-game tournament in 2016, a computer program dubbed AlphaGo created by the company DeepMind triumphed over world champion Lee Sedol. Deep neural networks and Monte Carlo tree search were combined in AlphaGo to analyze potential moves and decide on a course of action.

Other Go-playing software has since been created, some of which is even more sophisticated than AlphaGo. To enhance their gaming abilities, these systems employ a variety of strategies including reinforcement learning, self-play, and neural networks. Some of these computers are even able to pick up strategies from human games and create their own special playing techniques.

In addition to pushing the limits of AI research, the creation of sophisticated Go-playing systems has had a big impact on the game itself. Many professional players have begun utilizing these programs to hone their own skills, and some have even adopted the playing philosophies of the programs. As a result, a new era of Go has begun, in which human and artificial intelligence (AI) players work together to advance the game.

Go is a difficult and intricate game that has traditionally been thought to be beyond the computational power of computers. But thanks to recent developments in AI, there are now tools that can play Go at a high level. These systems have already had a huge impact on the game and on AI research as a whole, despite the fact that they are not yet flawless and still have much to learn. It is likely that Go-playing algorithms will continue to advance in the years to come, and it will be fascinating to see how this will alter both the game and our perception of AI.