Historic Milestone: AlphaZero

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Rebel, chrisw

Daniel Shawul
Posts: 4185
Joined: Tue Mar 14, 2006 11:34 am
Location: Ethiopia

Re: MCTS-NN vs alpha-beta

Post by Daniel Shawul »

Thanks Albert! Well written article.

A minor comment is that there is a distinction b/n AlphaGo Zero, and AlphaZero but I guess you left this out deliberately to keep it simple.
Maybe Google had this chess results when they first published AlphaGo Zero anyway -- only been a few months b/n these two.

Daniel
User avatar
fern
Posts: 8755
Joined: Sun Feb 26, 2006 4:07 pm

Re: Much weaker than Stockfish

Post by fern »

" just that outcalculates Stockfish consistently...."

Well, that is enough, pal.

Fern
Leo
Posts: 1076
Joined: Fri Sep 16, 2016 6:55 pm
Location: USA/Minnesota
Full name: Leo Anger

Re: Historic Milestone: AlphaZero

Post by Leo »

Lion wrote:While I understand why you would like to do that (so would I), I believe it will never happen.
This is surely not the original intend of the team that worked on AlphaZero.

They just proved that by letting an IA develop within a few hours it could get more than a match for a regular program on which several thousands of hours were spent....

Unfortunately, the proof is done and they will likely move to the next things. SF with opening book, this or that..... is most probably of little interest to them.

rgds
Agree.
Advanced Micro Devices fan.
Albert Silver
Posts: 3019
Joined: Wed Mar 08, 2006 9:57 pm
Location: Rio de Janeiro, Brazil

Re: Historic Milestone: AlphaZero

Post by Albert Silver »

fern wrote:In a way this is the end of chess.
In 25 years, the end of us as the leading mental entity in the earth.

Becoming a kitten regards
F
I'll be honest, I don't get these comments at all. AlphaZero played great, no question, and the games showed a computer engine that did nt suffer from the same problems we usually see. This too was great.

However, the end result was still a 102 Elo performance over SF on a 32-core machine. I never thought to myself: "if we just get 102 more Elo on Stockfish it will be the end of chess."
"Tactics are the bricks and sticks that make up a game, but positional play is the architectural blueprint."
User avatar
Laskos
Posts: 10948
Joined: Wed Jul 26, 2006 10:21 pm
Full name: Kai Laskos

Re: Historic Milestone: AlphaZero

Post by Laskos »

fern wrote:In a way this is the end of chess.
In 25 years, the end of us as the leading mental entity in the earth.

Becoming a kitten regards
F
Fern, be kitten, a silly kitten, and drink wine, good wine! No one will ever beat us at that!

Cheers pal!
Leo
Posts: 1076
Joined: Fri Sep 16, 2016 6:55 pm
Location: USA/Minnesota
Full name: Leo Anger

Re: MCTS-NN vs alpha-beta

Post by Leo »

Albert Silver wrote:
Rebel wrote:Can't believe it without a press release from Google.
They did more than that. They published a scientific paper anyone can read.

https://en.chessbase.com/post/the-futur ... arns-chess
Albert, you wrote a nice article. Thanks.
Advanced Micro Devices fan.
Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: Much weaker than Stockfish

Post by Lyudmil Tsvetkov »

MikeGL wrote:
Lyudmil Tsvetkov wrote:
MikeGL wrote:
Lyudmil Tsvetkov wrote:
MikeGL wrote:
Lyudmil Tsvetkov wrote:
MikeGL wrote:
Lyudmil Tsvetkov wrote:The training matches are different from the 100 games match with 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.
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?
12 opennings x 100 = 1,200 games total.
Before we were talking about 300 and 100, now 1200 suddenly appears...
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.
Read the series of posts properly. It is 100 games per openning, you clearly don't understand Table 2.

300 games because you were talking about 1.e4 earlier which appears in 6 diagrams.
How many is 50 x 6 ?
You were claiming that AlphaZero didn't play 1.e4, i told you it did! 300 times it did play 1.e4 against SF8.

See the total summation below: 1,200 games for all 12 opennings. Come on man, even this very
basic stuff we argue?
Do you have the pgn for the training games, which, btw., are claimed to run into the thousands?
Note that all Training games are self-play (no SF8 involved). The 1,200 are all match games against SF8.
No data given in PDF about the total number of self-play for learning, neither were the self-play PGN published. Only SF8 match was published.
100 game match versus SF8 was played on all 12 common ECO openings.
I guess you are confused of the plotted graph being put beside the
diagram. The graph is self-play, the diagram is SF8 match.
Try to read the caption of Table 2 properly. 37 times if need be.
100%12=4, so there is a remainder, and this is fully absurd.
1,200 games it was, read the table correctly. You are insisting total games were only 100, yes 100 for EACH openning, but there are 12 opennings so 12 times 100 is 1,200.
Are you serious? This is pre-school basic math
That is what you have written: 100 game match versus SF8 was played on all 12 common ECO openings.
kranium
Posts: 2129
Joined: Thu May 29, 2008 10:43 am

Re: Much weaker than Stockfish

Post by kranium »

Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: Much weaker than Stockfish

Post by Lyudmil Tsvetkov »

MikeGL wrote:
Lyudmil Tsvetkov wrote:
KWRegan wrote:
Lyudmil Tsvetkov wrote:Why don't they disclose what their evaluation is: that will be a big step towards knowing the truth.
They can't. The evaluation is a sequence of numbers specifying myriad weights on umpteen-dozen layers of a neural network. This aspect (of the original AlphaGo) in contrast to Stockfish is addressed in my Feb. 2016 article https://rjlipton.wordpress.com/2016/02/07/magic-to-do/ That this is endemic to "deep learning" has energized a counter-push toward "Explainable AI."


What I wish to know better, incidentally, is the memory footprint of their trained network and how portable it is.
They are still tuning at the level of a 2850 single core engine, so things will just get significantly more difficult in the future, when the quality of the terms will have much higher impact.
Mentioned in the paper, the eval is non-linear, not like the current engines that
uses linear eval functions. They are not tuning the eval, the AI itself is tuning the eval
autonomously without human input.
You wanna bet that they don't change the code?
No human interference at all?
Why do they need all that big team then?

There is no concept like artificial intelligence, there is simply no such concept, someone conceived it wrong and it spread across the world.
The machine is just executing the code written by humans, and it will always be so. It does not matter at all if it is linear or non-linear, it is still executing the code.
And that code changes. Basically, a huge autotuner, with what are CLOP and other tuners different, essentially?

For example, it should have started from somewhere, they should have at least the basic piece values and some psqt values. Then, when it wins a game and an e5 pawn is featured or an advanced knight outpost, the autotuner will increase their values, so what? How advanced is that?
What is the difference with normal tuning, I just can't see it?

No machine can start without a code, and that code will guide all its operations, EVER, regardless of whether it is a multilayer algorithm or not.

It is you who are mixing up things, not me.

You also believe they improved so much in 4 hours?
Want to bet they will not have a 5000 elo engine in another week, do you?
If they don't, then why are you arguing?

I have been there mentally and I know what is necessary to construct a much stronger chess playing entity. They don't have the necessary preconditions and will never succeed. Still 1850 currently.
Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: Historic Milestone: AlphaZero

Post by Lyudmil Tsvetkov »

Spliffjiffer wrote:personally i have no doubt that the "4 hours thing" is the truth...and even if its 4 days, thats no difference imo....the fact, that that "process" manages to get to such a ultra-high level of playingstrengh is nothing other than revolutionary and thats simply a scary, very scary, fact that we are faced with :-(
But if in a week's time it is not 1000 elo stronger, you will repent. :)