Previous amazing Alpha Zero position is not amazing anymore!

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

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
shrapnel
Posts: 1261
Joined: Fri Nov 02, 2012 8:43 am
Location: New Delhi, India

Re: Previous amazing Alpha Zero position is not amazing anymore!

Post by shrapnel » Mon Nov 19, 2018 8:26 am

peter wrote:
Mon Nov 19, 2018 7:34 am
And what makes you think, A0 could find (or could have found, I doubt it exists anymore) all tactical difficult best moves, that SF finds immediately, at all or at least in a reasonable time
The FACT that AlphaZero WIPED the floor with Stockfish; not just beat Stockfish, but HUMILIATED Stockfish.
If you go over the games,( see YouTube) its easy to make out that Stockfish didn't have a CLUE about what was going on !
You can run your Tactical Tests or whatever till the Cows come home, but NOTHING will change the fact that Stockfish was thoroughly out-classed over the Board by an Engine that was playing at a totally different, stratospheric level.
In simple words, AlphaZero made Stockfish look STUPID !
Stockfish has been beaten many times by a Komodo/Houdini, but NEVER in this humiliating fashion.
THAT is what is AWESOME !
i7 5960X @ 4.1 Ghz, 64 GB G.Skill RipJaws RAM, Twin Asus ROG Strix OC 11 GB Geforce 2080 Tis

Morten Lohne
Posts: 9
Joined: Wed Jul 19, 2017 10:18 pm

Re: Previous amazing Alpha Zero position is not amazing anymore!

Post by Morten Lohne » Mon Nov 19, 2018 2:29 pm

shrapnel wrote:
Mon Nov 19, 2018 8:26 am
peter wrote:
Mon Nov 19, 2018 7:34 am
And what makes you think, A0 could find (or could have found, I doubt it exists anymore) all tactical difficult best moves, that SF finds immediately, at all or at least in a reasonable time
The FACT that AlphaZero WIPED the floor with Stockfish; not just beat Stockfish, but HUMILIATED Stockfish.
If you go over the games,( see YouTube) its easy to make out that Stockfish didn't have a CLUE about what was going on !
You can run your Tactical Tests or whatever till the Cows come home, but NOTHING will change the fact that Stockfish was thoroughly out-classed over the Board by an Engine that was playing at a totally different, stratospheric level.
In simple words, AlphaZero made Stockfish look STUPID !
Stockfish has been beaten many times by a Komodo/Houdini, but NEVER in this humiliating fashion.
THAT is what is AWESOME !
You need to calm down a bit. They released 10 pgns played by a poorly configured Stockfish on mostly unknown hardware vs AlphaZero on mostly unknown hardware. Again, they only released 10 handpicked pgns. You can choose to carried away be the hype, or choose to accept that not many conclusions can be drawn from it at all.

shrapnel
Posts: 1261
Joined: Fri Nov 02, 2012 8:43 am
Location: New Delhi, India

Re: Previous amazing Alpha Zero position is not amazing anymore!

Post by shrapnel » Tue Nov 20, 2018 7:18 pm

Morten Lohne wrote:
Mon Nov 19, 2018 2:29 pm
You need to calm down a bit. They released 10 pgns played by a poorly configured Stockfish on mostly unknown hardware vs AlphaZero on mostly unknown hardware. Again, they only released 10 handpicked pgns. You can choose to carried away be the hype, or choose to accept that not many conclusions can be drawn from it at all.
Yes, the lack of interest shown by the Google Team which created Alphazero which went on to beat Stockfish with contemptuous ease, works out in your and the other nay sayers favor.
Disappointingly, Google showed no interest in Commercializing AlphaZero and answering the Critics and went on to other things.
For them it was just an exercise to show the value of AI in beating Go and computer chess engines over which their Developers and Testers had spent nearly their entire lives.
Plain and simple truth is that AlphaZero made a Mockery of the alpha-beta Computer Chess Engines over which many Developers and Testers spend almost their entire lives (unbearable I know).
Anyway, good for Komodo Team and Houdart, who would otherwise have to look for new jobs.
i7 5960X @ 4.1 Ghz, 64 GB G.Skill RipJaws RAM, Twin Asus ROG Strix OC 11 GB Geforce 2080 Tis

User avatar
Eduard
Posts: 243
Joined: Fri Oct 26, 2018 10:58 pm
Location: Germany
Full name: Eduard Nemeth
Contact:

Re: Previous amazing Alpha Zero position is not amazing anymore!

Post by Eduard » Wed Nov 21, 2018 2:25 am

On my Smartphone same effect!

Komodo 11.3.1 with Dynamism 325, finds the move after few seconds!

Gian-Carlo Pascutto
Posts: 1192
Joined: Sat Dec 13, 2008 6:00 pm
Contact:

Re: Previous amazing Alpha Zero position is not amazing anymore!

Post by Gian-Carlo Pascutto » Wed Nov 21, 2018 8:21 am

shrapnel wrote:
Tue Nov 20, 2018 7:18 pm
Disappointingly, Google showed no interest in Commercializing AlphaZero
They just announced a book with 2000 extra games to be published by New In Chess.

You can go buy it and gasp at the games then.
Plain and simple truth is that AlphaZero made a Mockery of the alpha-beta Computer Chess Engines over which many Developers and Testers spend almost their entire lives (unbearable I know).
Only in the eyes of people that don't understand how fast Stockfish has improved, what timecontrols or hashtables are, etc. The only thing that annoys me is that their marketing is so effective in suspending people's critical thought. I feel like the latter is going to come back and haunt us someday outside the chess world. Or maybe it already is.

shrapnel
Posts: 1261
Joined: Fri Nov 02, 2012 8:43 am
Location: New Delhi, India

Re: Previous amazing Alpha Zero position is not amazing anymore!

Post by shrapnel » Wed Nov 21, 2018 10:30 am

They just announced a book with 2000 extra games to be published by New In Chess.

You can go buy it and gasp at the games then.
Really ? I certainly will.
Link please.
what timecontrols or hashtables are
As far as I know, both Stockfish and AlphaZero had 1 minute for each move.
Losers whiners.
i7 5960X @ 4.1 Ghz, 64 GB G.Skill RipJaws RAM, Twin Asus ROG Strix OC 11 GB Geforce 2080 Tis

Uri Blass
Posts: 8650
Joined: Wed Mar 08, 2006 11:37 pm
Location: Tel-Aviv Israel

Re: Previous amazing Alpha Zero position is not amazing anymore!

Post by Uri Blass » Wed Nov 21, 2018 10:34 am

shrapnel wrote:
Mon Nov 19, 2018 8:26 am
peter wrote:
Mon Nov 19, 2018 7:34 am
And what makes you think, A0 could find (or could have found, I doubt it exists anymore) all tactical difficult best moves, that SF finds immediately, at all or at least in a reasonable time
The FACT that AlphaZero WIPED the floor with Stockfish; not just beat Stockfish, but HUMILIATED Stockfish.
If you go over the games,( see YouTube) its easy to make out that Stockfish didn't have a CLUE about what was going on !
You can run your Tactical Tests or whatever till the Cows come home, but NOTHING will change the fact that Stockfish was thoroughly out-classed over the Board by an Engine that was playing at a totally different, stratospheric level.
In simple words, AlphaZero made Stockfish look STUPID !
Stockfish has been beaten many times by a Komodo/Houdini, but NEVER in this humiliating fashion.
THAT is what is AWESOME !
The facts are that there was no fair competition between alphazero and stockfish.

Stockfish did not use the best hardware it can use.
Stockfish did not use an opening book.
Stockfish did not use time control that people use(nobody test stockfish at fixed time per move so they did not optimize it for these conditions)
Stockfish did not use tablebases
Stockfish did not use latest developement version at the time

The alphazero team did not tell the stockfish team that they want to play a 100 game match against stockfish so the stockfish team care to prepare stockfish for the match.

shrapnel
Posts: 1261
Joined: Fri Nov 02, 2012 8:43 am
Location: New Delhi, India

Re: Previous amazing Alpha Zero position is not amazing anymore!

Post by shrapnel » Wed Nov 21, 2018 10:54 am

Again, Losers Whiners.
Nor was AlphaZero using an Opening Book or Tablebases.
As for quibbles like Time Control Tuning and latest Dev Version, you would do well to keep in mind that AlphaZero was given just 4 Hours Training before the Match and before 4 Hours didn't even know the chess moves !
Anyway, no reasoning with the Flat-Earth believers.
Here's a link to an interesting article on AlphaZero..
https://www.nextbigfuture.com/2018/01/c ... style.html
Interesting quote from that Article is "If Karpov had been a chess engine, he might have been called AlphaZero. There is a relentless positional boa constrictor approach that is simply unheard of. Modern chess engines are focused on activity, and have special safeguards to avoid blocked positions as they have no understanding of them and often find themselves in a dead end before they realize it. AlphaZero has no such prejudices or issues, and seems to thrive on snuffing out the opponent’s play. It is singularly impressive, and what is astonishing is how it is able to also find tactics that the engines seem blind to."
i7 5960X @ 4.1 Ghz, 64 GB G.Skill RipJaws RAM, Twin Asus ROG Strix OC 11 GB Geforce 2080 Tis

duncan
Posts: 10850
Joined: Mon Jul 07, 2008 8:50 pm

Re: Previous amazing Alpha Zero position is not amazing anymore!

Post by duncan » Wed Nov 21, 2018 11:41 am

shrapnel wrote:
Wed Nov 21, 2018 10:30 am
They just announced a book with 2000 extra games to be published by New In Chess.

You can go buy it and gasp at the games then.
Really ? I certainly will.
Link please.
what timecontrols or hashtables are
As far as I know, both Stockfish and AlphaZero had 1 minute for each move.
Losers whiners.

https://www.newinchess.com/game-changer

Expected March 2019

It took AlphaZero only a few hours of self-learning to become the chess player that shocked the world.

The artificial intelligence system, created by DeepMind, had been fed nothing but the rules of the Royal Game when it beat the world’s strongest chess engine in a prolonged match. The selection of ten games published in December 2017 created a worldwide sensation: how was it possible to play in such a brilliant and risky style and not lose a single game against an opponent of superhuman strength?

For Game Changer, Matthew Sadler and Natasha Regan investigated more than two thousand previously unpublished games by AlphaZero. They also had unparalleled access to its team of developers and were offered a unique look ‘under the bonnet’ to grasp the depth and breadth of AlphaZero’s search. Sadler and Regan reveal its thinking process and tell the story of the human motivation and the techniques that created AlphaZero.

Game Changer also presents a collection of lucidly explained chess games of astonishing quality. Both professionals and club players will improve their game by studying AlphaZero’s stunning discoveries in every field that matters: opening preparation, piece mobility, initiative, attacking techniques, long-term sacrifices and much more.

...


Matthew Sadler (1974) is a Grandmaster who twice won the British Championship and was awarded an individual Gold Medal at the 1996 Olympiad. He has authored several highly acclaimed books on chess and has been writing the famous ‘Sadler on Books’ column for New In Chess magazine for many years. Natasha Regan is a Women’s International Master from England who achieved a degree in mathematics from Cambridge University. Matthew Sadler and Natasha Regan won the English Chess Federation 2016 Book of the Award for their book Chess for Life.

jp
Posts: 889
Joined: Mon Apr 23, 2018 5:54 am

Re: Previous amazing Alpha Zero position is not amazing anymore!

Post by jp » Wed Nov 21, 2018 11:54 am

duncan wrote:
Wed Nov 21, 2018 11:41 am
https://www.newinchess.com/game-changer

For Game Changer, Matthew Sadler and Natasha Regan investigated more than two thousand previously unpublished games by AlphaZero. They also had unparalleled access to its team of developers and were offered a unique look ‘under the bonnet’ to grasp the depth and breadth of AlphaZero’s search. Sadler and Regan reveal its thinking process and tell the story of the human motivation and the techniques that created AlphaZero.
All they need to show is all of A0's games, not just good-looking ones. No "stories", no "unique looks", no "thinking processes". Just ALL of the games. It's that simple.

Post Reply