The computer program that defeated the world Go champion taught itself how to improve its game by playing millions of matches against itself, according to the head of the Google subsidiary that developed the software. “In AlphaGo’s case, that involved splitting itself in half and playing millions of matches against itself, learning from each victory and loss. In one day alone, AlphaGo was able to play itself more than a million times, gaining more practical experience than a human player could hope to gain in a lifetime. In essence, AlphaGo got better at Go simply by thinking extremely hard about the problem.” Read more