A neural net system built by Google has beaten the European champion in the Chinese game of Go, winning five out of five games and crossing a new threshold for machine intelligence.
In a recent paper published in Nature, DeepMind researchers revealed how the system was constructed and how it was able to succeed where decades of previous Go systems have failed. Go has long been considered one of the hardest games to automate, making the new DeepMind system particularly interesting for artificial intelligence researchers. 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. Here we introduce a new approach to computer Go that uses ‘value networks’ to evaluate board positions and ‘policy networks’ to select moves.
They used a neural networks trained by combinations of supervised learning from human expert games, and reinforcement learning from games of self-play. “Without any lookahead search, the neural networks play Go at the level of state-of-the-art Monte Carlo tree search programs that simulate thousands of random games of self-play”, they said.