Deep Mind Co-Founder and CEO Demis Hassabis has revealed that the updated version of AlphaGo was the AI behind the dozens of high-profile Go online game wins in late December last year.
The updated version of AlphaGo, playing under the alias Master(P), won more than 50 Go games in late December 2016 on the online servers Tygem and FoxGo. Nature News reported that out of over 50 games played by Master(P), only one game was considered a tie because the network connection of Master(P)’s opponent, Go professional player Chen Yaoye, timed out. The updated version of AlphaGo also defeated 19-year-old Ke Jie, the world number one Go online player, and Go great grandmaster Gu Li.
The CEO of DeepMind tweeted, “We’ve been hard at work improving AlphaGo, and over the past few days we’ve played some unofficial online games at fast time controls with our new prototype version, to check that it’s working as well as we hoped.”
— Demis Hassabis (@demishassabis) January 4, 2017
Go is considered as the most complex game created by humans. AlphaGo – an AI created by DeepMind, Alphabet’s London-based AI firm – caught the public attention when it defeated the number one Go professional player Lee Sedol, winning 4 out of 5 games in March 2016.
After losing to Master(P), Gu Li offered a reward of $14,400 (100,000 yuan) to anyone who could beat the mystery player – who turned out to be the updated version of AlphaGo. After the big reveal of AlphaGo, Gu Li said, “Together, humans and AI will soon uncover the deeper mysteries of Go.”
As to the future plans of the updated version of AlphaGo, Hassabis said, “Now that our official testing is complete, we’re looking forward to playing some official, full-length games later this year in collaboration with Go organizations and experts, to explore the profound mysteries of the game further in this spirit of mutual enlightenment.”