4 ms·

AlphaZero and Stockfish did not run on the same hardware. So it's not clear if the algorithm is better or the algorithm was just run faster.

seanwilson 9y ago
Given the algorithm didn't require tuning and it examined less moves each turn I would say it was a better algorithm. Lots of AI problems aren't solved with current techniques just by throwing more CPU power at them so it would still be impressive even if it required a lot of computing power.
- dzdt 9y ago
  The thing it does require is specialized hardware (TPU) to run at a reasonable speed.
  - pixl97 9y ago
    A TPU isn't specialized hardware in the same way a CPU or GPU are not task specialized. All three are generic execution platforms that are optimized for a particular type of processing, but do so in a generalized fashion that can complete many different types of tasks.
animal531 9y ago
From the pdf: "AlphaZero searches just 80 thousand positions per second in chess...compared to 70 million for Stockfish" But even that isn't the important part in my opinion. They basically created an ML solution that can learn to play these various games, in very little time (hours), while beating other previous engines (by decent to ridiculous margins). They've created a more generalised solution to their previous AlphaGo engine, which may be very useful in future.
- no_gravity 9y ago
  'AlphaZero searches just 80 thousand positions per second' As I understand it, we don't really know what AZ does when it evaluates a position. As it was not explicitly programmed. It could do something that is similar to evaluating more positions.
  - tmalsburg2 9y ago
    I was assuming that AZ is using a tree search strategy similar to conventional chess engines but with a neural network as a more sophisticated board evaluation function. If true (is it?) you can tell how many positions it evaluates per unit of time.
    - sanxiyn 9y ago
      No, AZ does not use tree search similar to conventional chess engines. That's an actual surprise. Neural network is used for two things: evaluation, yes, but also much more importantly, search selectivity. In AlphaGo Zero paper, they show that selectivity is so important that playing solely from selectivity (that is, ask neural network which move one should search first, and play that move without searching at all) results in professional level, see Figure 6b. Fan Hui level, not Lee Sedol level, but still.
      - kilburn 9y ago
        That is game-dependent, so we can't be sure it would result in pro-level when playing chess. In fact, it is very possible that it wouldn't because chess has a much smaller branching factor than go (and many more practically forced moves etc.). Also, changing the heuristic you use to chose candidates (selectivity) doesn't mean you're not doing search anymore!
      - readams 9y ago
        they both use a tree search, though it's a different tree search algorithm.
edraferi 9y ago
Also, Stockfish was denied some initialization data that it usually uses: The player with most strident objections to the conditions of the match was GM Hikaru Nakamura. While a heated discussion is taking place online about processing power of the two sides, Nakamura thought that was a secondary issue. The American called the match "dishonest" and pointed out that Stockfish's methodology requires it to have an openings book for optimal performance. While he doesn't think the ultimate winner would have changed, Nakamura thought the size of the winning score would be mitigated.
- sanxiyn 9y ago
  While I somewhat sympathize with Nakamura in that in case of AlphaGo vs Lee Sedol, Lee certainly had an "opening book", I disagree with your characterization that opening book is "usual". TCEC is widely recognized competition for chess engines and TCEC rule specifies no opening book. For engine-to-engine evaluation, no opening book is the usual method of evaluation. One problem is that in a sense AlphaZero has an "opening book" encoded in its neural network weights. But just like it is unclear how to construct "Lee Sedol without an opening book" at all, it is unclear how to construct "AlphaZero without an opening book in such sense". So indeed, while unusual for engine evaluation, it probably is best to play against Stockfish with an opening book.
  - sanderjd 9y ago
    Seems like an easy answer to this would be to just let Stockfish use its opening book?
    - sanxiyn 9y ago
      That's the problem. Stockfish has no opening book. While there is no doubt Stockfish can play stronger with good opening book, keeping a book up to date with engine changes is really a full time job. So Stockfish project does not have any official opening book. Personally, I think using publicly available Komodo book would have been enough, but obviously Komodo book is tuned for Komodo and everybody would complain about any book problem. In a sense, "no opening book" is the official upstream supported configuration, so it is entirely a defensible choice.
grondilu 9y ago
Still, when you look at the games, it's hard not to think something genuinely new has happened. Several grandmasters have expressed amazement at the style of play. AlphaZero won some games in a romantic style that's reminiscent of old champions. It's definitely not the kind of play we've been accustomed to with chess engines : AlphaZero seems to rely on a deep strategic understanding of piece placement and dynamism opportunities. It's difficult to attribute some of these wins to a simple superiority in computing power.
- espadrine 9y ago
  > AlphaZero won some games in a romantic style The difference in style is likely influenced by the insertion of historical boards as input to the neural network. The sequence of moves are therefore more likely to look related to one another.
  - grondilu 9y ago
    The neural network was only fed with games from self-play. No historical game whatsoever was given. At least that's Deepmind's claim.
    - espadrine 9y ago
      I was misunderstood. By historical, I mean, it includes the past N boards, which pushes the network to make actions correlated to the previous actions performed. That is similar to how humans play.
- no_gravity 9y ago
  Imagine a computer with a simple brute force algorithm but unlimited computing power. It would win 100:0 against AZ. Would the grandmasters look at the games and say "Yes, it won every time but it's style is rather clumsy"?
  - grondilu 9y ago
    There is no unlimited computing power anywhere, though. Not even in Google's quarters. The style difference is so profound that it is difficult to impute computing power only, because we know what kind of improvement reasonable computing power gives : it makes the engine stronger but does not quite change its style.

seanwilson 9y ago

Given the algorithm didn't require tuning and it examined less moves each turn I would say it was a better algorithm. Lots of AI problems aren't solved with current techniques just by throwing more CPU power at them so it would still be impressive even if it required a lot of computing power.

dzdt 9y ago
The thing it does require is specialized hardware (TPU) to run at a reasonable speed.
- pixl97 9y ago
  A TPU isn't specialized hardware in the same way a CPU or GPU are not task specialized. All three are generic execution platforms that are optimized for a particular type of processing, but do so in a generalized fashion that can complete many different types of tasks.

animal531 9y ago

From the pdf: "AlphaZero searches just 80 thousand positions per second in chess...compared to 70 million for Stockfish" But even that isn't the important part in my opinion. They basically created an ML solution that can learn to play these various games, in very little time (hours), while beating other previous engines (by decent to ridiculous margins). They've created a more generalised solution to their previous AlphaGo engine, which may be very useful in future.

no_gravity 9y ago
'AlphaZero searches just 80 thousand positions per second' As I understand it, we don't really know what AZ does when it evaluates a position. As it was not explicitly programmed. It could do something that is similar to evaluating more positions.
- tmalsburg2 9y ago
  I was assuming that AZ is using a tree search strategy similar to conventional chess engines but with a neural network as a more sophisticated board evaluation function. If true (is it?) you can tell how many positions it evaluates per unit of time.
  - sanxiyn 9y ago
    No, AZ does not use tree search similar to conventional chess engines. That's an actual surprise. Neural network is used for two things: evaluation, yes, but also much more importantly, search selectivity. In AlphaGo Zero paper, they show that selectivity is so important that playing solely from selectivity (that is, ask neural network which move one should search first, and play that move without searching at all) results in professional level, see Figure 6b. Fan Hui level, not Lee Sedol level, but still.
    - kilburn 9y ago
      That is game-dependent, so we can't be sure it would result in pro-level when playing chess. In fact, it is very possible that it wouldn't because chess has a much smaller branching factor than go (and many more practically forced moves etc.). Also, changing the heuristic you use to chose candidates (selectivity) doesn't mean you're not doing search anymore!
    - readams 9y ago
      they both use a tree search, though it's a different tree search algorithm.

edraferi 9y ago

Also, Stockfish was denied some initialization data that it usually uses: The player with most strident objections to the conditions of the match was GM Hikaru Nakamura. While a heated discussion is taking place online about processing power of the two sides, Nakamura thought that was a secondary issue. The American called the match "dishonest" and pointed out that Stockfish's methodology requires it to have an openings book for optimal performance. While he doesn't think the ultimate winner would have changed, Nakamura thought the size of the winning score would be mitigated.

sanxiyn 9y ago
While I somewhat sympathize with Nakamura in that in case of AlphaGo vs Lee Sedol, Lee certainly had an "opening book", I disagree with your characterization that opening book is "usual". TCEC is widely recognized competition for chess engines and TCEC rule specifies no opening book. For engine-to-engine evaluation, no opening book is the usual method of evaluation. One problem is that in a sense AlphaZero has an "opening book" encoded in its neural network weights. But just like it is unclear how to construct "Lee Sedol without an opening book" at all, it is unclear how to construct "AlphaZero without an opening book in such sense". So indeed, while unusual for engine evaluation, it probably is best to play against Stockfish with an opening book.
- sanderjd 9y ago
  Seems like an easy answer to this would be to just let Stockfish use its opening book?
  - sanxiyn 9y ago
    That's the problem. Stockfish has no opening book. While there is no doubt Stockfish can play stronger with good opening book, keeping a book up to date with engine changes is really a full time job. So Stockfish project does not have any official opening book. Personally, I think using publicly available Komodo book would have been enough, but obviously Komodo book is tuned for Komodo and everybody would complain about any book problem. In a sense, "no opening book" is the official upstream supported configuration, so it is entirely a defensible choice.

grondilu 9y ago

Still, when you look at the games, it's hard not to think something genuinely new has happened. Several grandmasters have expressed amazement at the style of play. AlphaZero won some games in a romantic style that's reminiscent of old champions. It's definitely not the kind of play we've been accustomed to with chess engines : AlphaZero seems to rely on a deep strategic understanding of piece placement and dynamism opportunities. It's difficult to attribute some of these wins to a simple superiority in computing power.

espadrine 9y ago
> AlphaZero won some games in a romantic style The difference in style is likely influenced by the insertion of historical boards as input to the neural network. The sequence of moves are therefore more likely to look related to one another.
- grondilu 9y ago
  The neural network was only fed with games from self-play. No historical game whatsoever was given. At least that's Deepmind's claim.
  - espadrine 9y ago
    I was misunderstood. By historical, I mean, it includes the past N boards, which pushes the network to make actions correlated to the previous actions performed. That is similar to how humans play.
no_gravity 9y ago
Imagine a computer with a simple brute force algorithm but unlimited computing power. It would win 100:0 against AZ. Would the grandmasters look at the games and say "Yes, it won every time but it's style is rather clumsy"?
- grondilu 9y ago
  There is no unlimited computing power anywhere, though. Not even in Google's quarters. The style difference is so profound that it is difficult to impute computing power only, because we know what kind of improvement reasonable computing power gives : it makes the engine stronger but does not quite change its style.