Last August, Microsoft Research Asia detailed an AI system dubbed Super Phoenix (Suphx for short) that could defeat Mahjong players after learning from only 5,000 matches. A revised preprint paper out this week delves a bit deeper, revealing that Suphx -- whose performance improved with additional training -- is now rated above 99.99% of all ranked human players on Tenhou, a Japan-based global online Mahjong competition platform with over 350,000 members. Building superhuman programs for games is a longstanding goal of the AI research community -- and not without good reason. Games are an analog of the real world, with a measurable objective, and they can be played an infinite amount of times across hundreds (or thousands) of powerful machines. Moreover, its researchers assert that the learnings are applicable to other domains, like the enterprise, where mundane but cognitively demanding tasks impact workers' productivity.

Artificial Intelligence (AI) has achieved great success in many domains, and game AI is widely regarded as its beachhead since the dawn of AI. In recent years, studies on game AI have gradually evolved from relatively simple environments (e.g., perfect-information games such as Go, chess, shogi or two-player imperfect-information games such as heads-up Texas hold'em) to more complex ones (e.g., multi-player imperfect-information games such as multi-player Texas hold'em and StartCraft II). Mahjong is a popular multi-player imperfect-information game worldwide but very challenging for AI research due to its complex playing/scoring rules and rich hidden information. We design an AI for Mahjong, named Suphx, based on deep reinforcement learning with some newly introduced techniques including global reward prediction, oracle guiding, and run-time policy adaptation. Suphx has demonstrated stronger performance than most top human players in terms of stable rank and is rated above 99.99% of all the officially ranked human players in the Tenhou platform. This is the first time that a computer program outperforms most top human players in Mahjong.

Artificial intelligence has thrashed humans at chess. Now the bots are gunning for mahjong. An AI-powered program developed by Microsoft Corp has surpassed the average level of the top players in a recent competition in Japan, Harry Shum, executive vice-president of the companys artificial intelligence and research group, said in Shanghai on Thursday. To those friends who usually lose money in mahjong, this is good news to you, Shum said to laughter at the World AI Conference. The bot player developed by Microsoft can deal with high uncertainty, presenting instincts akin to human, projection and deduction capabilities as well as a sense of overall consciousness.

In 1952, Professor Sandy Douglas created a tic-tac-toe game on the EDSAC, a room-sized computer at the University of Cambridge. One of the first ever computer games, it was developed as part of a thesis on human-computer interaction. Forty-five years later, in 1997, another milestone occurred when IBM's Deep Blue machine defeated Garry Kasparov, the world chess champion. This was followed by Watson, again created by IBM, which appeared on the Jeopardy! Yet another breakthrough was Google's DeepMind AlphaGo, which in 2016 defeated the Go world champion Lee Se-dol at a tournament in South Korea.

The Rivers Casino in Pittsburgh may not seem a likely setting for a major scientific breakthrough. But on Tuesday, it was: Libratus, an AI system developed by Carnegie Mellon University, beat the world's top four human players in a 20-day tournament of Head's-Up No-Limit Texas Hold'em poker. Libratus, developed by Carnegie Mellon's Tuomas Sandholm, a professor of computer science, and Noam Brown, a Ph.D. student in computer science, competed against Dong Kim, Jimmy Chou, Daniel McAulay, and Jason Les in a competition called "Brains Vs. Artificial Intelligence: Upping the Ante"--during which 120,000 hands were played. "This is the last frontier," said Sandholm during a press conference on Tuesday.

Artificial intelligence researchers at DeepMind, a research unit that operates under the umbrella, just surpassed a very significant benchmark in machine learning. It was first reported on Wednesday in Nature that the artificial intelligence system dubbed AlphaGo successfully beat a professional Go player, Fan Hui, in series of five matches. Go, a complex Chinese board game played by two players with black and white stones, has traditionally been deemed an excellent test for AI. According to a Google press release, previously computers have only been ale to play the game as successfully as amateur Go players, making this win all that much more significant. AlphaGo also won all but one of 500 matches against other top AI Go programs (a 99.8% success rate).