Well, probably lower than 3000 elos.jdart wrote:Even Stockfish on a Raspberry Pi is a strong chess player, though. So AlphaZero has decent performance, we just need a better comparison. I don't think there is any fundamental reason a NN based system such as AlphaZero couldn't run on commodity hardware, so maybe that can happen.
--Jon
Historic Milestone: AlphaZero
Moderators: hgm, Rebel, chrisw
-
- Posts: 6052
- Joined: Tue Jun 12, 2012 12:41 pm
Re: Historic Milestone: AlphaZero
-
- Posts: 6052
- Joined: Tue Jun 12, 2012 12:41 pm
Re: Much weaker than Stockfish
The training matches are different from the 100 games match with Stockfish.MikeGL wrote:Lyudmil Tsvetkov wrote:For some reason, I have no access to that table and the page as a whole.MikeGL wrote:I think Table 2 [ECO opennings] in the PDF would answer your argument.Lyudmil Tsvetkov wrote:
Alpha chooses only 1.d4 and 1.Nf3, while Stockfish goes for 1.e4
Judging from this, I can say that Alpha is much weaker than SF in terms of software, and the only reason for the win is the very big hardware advantage.
All those 12 common opennings (on that Table 2) was played by AlphaZero against SF8, 100 times each. and only a total of 4 losses (out of 300 games) as white starting with 1.e4 (for AlphaZero) as shown on that table.
Very weird.
It seems the page recognises its detractors.
The uppermost right diagram is a French Defence [C00], won't fit on my screen.
*
Just download the PDF (right-click then choose download) then upload again at gmail then view it from there.
Otherwise, just download and view it on your PC using any PDF reader.
It is not at all clear to me where were books used and where not.
12 openings with reversed colours don't square in any way with 100 played games, so did they actually left some openings played more than others, or did not they flip colours?
-
- Posts: 2129
- Joined: Thu May 29, 2008 10:43 am
Re: Much weaker than Stockfish
I'm sure opening books were not used...Lyudmil Tsvetkov wrote:
The training matches are different from the 100 games match with Stockfish.
It is not at all clear to me where were books used and where not.
12 openings with reversed colours don't square in any way with 100 played games, so did they actually left some openings played more than others, or did not they flip colours?
In the early self-play games things like 1.a3, 1.a4, etc. were probably tried by AlphaZero...
eventually it learned that 1. e4 or 1. d4 had the highest success rates.
-
- Posts: 546
- Joined: Sat Aug 17, 2013 12:36 am
Re: Much weaker than Stockfish
Time to place an order for one of these: https://www.nvidia.com/en-us/data-center/dgx-station/
Unless Google starts selling TPU2s, which I doubt.....
Unless Google starts selling TPU2s, which I doubt.....
-
- Posts: 4190
- Joined: Wed Nov 25, 2009 1:47 am
Re: Much weaker than Stockfish
If K80 is 5k$, that would be like 100k$ minimum starting price.jhellis3 wrote:Time to place an order for one of these: https://www.nvidia.com/en-us/data-center/dgx-station/
Unless Google starts selling TPU2s, which I doubt.....
-
- Posts: 1346
- Joined: Sat Apr 19, 2014 1:47 pm
Re: Historic Milestone: AlphaZero
I don't think any engine below 3000 elo would crush Stockfish with a strong hardware.Lyudmil Tsvetkov wrote:Well, probably lower than 3000 elos.jdart wrote:Even Stockfish on a Raspberry Pi is a strong chess player, though. So AlphaZero has decent performance, we just need a better comparison. I don't think there is any fundamental reason a NN based system such as AlphaZero couldn't run on commodity hardware, so maybe that can happen.
--Jon
-
- Posts: 1010
- Joined: Thu Sep 01, 2011 2:49 pm
Re: Much weaker than Stockfish
Yes, the plot on the diagram is the training game, but 100 games per openning was played. 50-50, and the score below the diagram is on AlphaZero perspective.Lyudmil Tsvetkov wrote:The training matches are different from the 100 games match with Stockfish.
12 opennings x 100 = 1,200 games total.12 openings with reversed colours don't square in any way with 100 played games, so did they actually left some openings played more than others, or did not they flip colours?
-
- Posts: 546
- Joined: Sat Aug 17, 2013 12:36 am
Re: Much weaker than Stockfish
It is $70,000 AFAIK.If K80 is 5k$, that would be like 100k$ minimum starting price.
-
- Posts: 6052
- Joined: Tue Jun 12, 2012 12:41 pm
Re: Much weaker than Stockfish
How can you be sure if they don't specify it?kranium wrote:I'm sure opening books were not used...Lyudmil Tsvetkov wrote:
The training matches are different from the 100 games match with Stockfish.
It is not at all clear to me where were books used and where not.
12 openings with reversed colours don't square in any way with 100 played games, so did they actually left some openings played more than others, or did not they flip colours?
In the early self-play games things like 1.a3, 1.a4, etc. were probably tried by AlphaZero...
eventually it learned that 1. e4 or 1. d4 had the highest success rates.
And it learned wrong. But 1.Nf3?
Is this engine still based on random choices? What perfect engine we are talking about then?
-
- Posts: 6052
- Joined: Tue Jun 12, 2012 12:41 pm
Re: Much weaker than Stockfish
Before we were talking about 300 and 100, now 1200 suddenly appears...MikeGL wrote:Yes, the plot on the diagram is the training game, but 100 games per openning was played. 50-50, and the score below the diagram is on AlphaZero perspective.Lyudmil Tsvetkov wrote:The training matches are different from the 100 games match with Stockfish.
12 opennings x 100 = 1,200 games total.12 openings with reversed colours don't square in any way with 100 played games, so did they actually left some openings played more than others, or did not they flip colours?
The 64/36 score certainly comes from 100 games, unless they assigned random points for a win.
And in that sample, I see Alpha playing just 1.d4 and 1.Nf3.