Many of AlphaZero's matches were also notable because it managed to achieve very clean strategic positions in a way that seem to be longer term and require more multitasking than a human game (AZs bishops, for example). While 'computer moves' are characterised by having a lack of 'conventional' logic to them, and instead relying on a non-intuitive or long-term combination. But over the course of a game there are usually one or two moves that are very much tells. Not to mention the fact that vast swathes of the opening are memorised computer moves. Top human players play so accurately, it would be almost impossible. I hope I was able to help clarify things with this short summary, and if you’re interested, I definitely recommend reading the paper itself for more in depth detail/information! You can find it freely available here at: Because as humans we don’t brute force possible moves, but rather immediately discount a large majority of plays and focus in on a very small subset that we know to be the optimal path to go towards, all through a sort of ingrained “intuition” built up over the countless games we’ve played. This is why AlphaZero performed far better than StockFish here given the same time, and one of the reasons why it was deemed to be more “human-like”. That’s the other benefit of training a model to learn things itself through self-play as opposed to manually engineering the features ourselves by hand. Stockfish also has its own value and pruning functions, but because of the fundamental nature of how expert systems work, it’s very difficult to create generalized features that address every possible variable. As even though both agents had exactly the same hardware and time allotted, AlphaZero is far more efficient at exploring its search space due to the value/pruning function it learned by itself through training. The primary achievement of what Alphazero represents, is that it was able to train itself completely independent of humans and come up entirely with its own features/strategies.Īt inference time too though, we see a large increase in benefit for AlphaZero against Stockfish. For the analogous comparison to how Stockfish works, it has been trained for more than a decade now with humans iteratively refining and manually creating features and strategies.
It’s important to make the distinction between training and inference time here though. I think what you might be mistakenly thinking of is, with some articles erroneously reporting when the games/paper were first publicly released, the fact that training a model of AlphaZero requires a considerable amount of self-play that takes up a lot of compute cycles.
There were no super computers involved there. At runtime, both the Stockfish and AlphaZero agents were given equivalent hardware, which was a standard commercial CPU.
0 kommentar(er)
