The Google DeepMind researchers have every reason to celebrate for creating history in the world of artificial intelligence. They reached a major breakthrough after 20 years of intense research, as they shaped the first machine to master the ancient Chinese game of Go.
Both FaceBook and Google were racing towards the same finishing line, but Google managed to beat the social network giant with its final product, AlphaGo. The program has been designed to play this classic game of artificial intelligence which has 10 to the power of 700 variations, and dates back to almost 2500 years in the history of Chinese civilization. The game beats the game of Chess hands down, which has only 10 to the power of 60 variations. "The game of Go has long been viewed as the most challenging of classic games for artificial intelligence owing to its enormous search space and the difficulty of evaluating board positions and moves." the researchers wrote in their paper, as per Outer Places.
Coincidentally, around the same time, Facebook also announced the launch of its similar program called Darkforest. Facebook CEO Mark Zuckerberg said that they were "getting close" to completing their research. According to Gizmodo, he also stated, "The researcher who works on this, Yuandong Tian, sits about 20 feet" from his desk, and added, "I love having our AI team right near me so I can learn from what they're working on."
The DeepMind researchers had adopted a unique approach for training the program - a combination of supervised observation for advanced levels of Go games and trial-and-error methods from self-play. The results were outstanding as the trained machine beat other Go machines at a 99.8% rate.
Following the completion of the training program, the AlphaGo team invited the three-time European Go Champion, Fan Hui, for a series of games, where, the machine not only matched the prowess of the human champion but beat him in all the five matches played a full-sized Go board.
"This is the first time that a computer Go program has defeated a human professional player, without handicap, in the full game of Go - a feat that was previously believed to be at least a decade away," said the DeepMind research paper, as represented in Business Insider.
With FaceBook dogging Google's footsteps, the breakthrough by the DeepMind team is meant to be a game changer mainly because it will have profound implications in the way artificial intelligence works, especially in areas of facial recognition abilities and predictive power. An average person is always looking for solutions by reasoning and logic via artificial intelligence. This innovative development will now pave the way for machines to move from one place to another through logical reasoning.
Join the Conversation