In news reminiscent of the initial AlphaZero shockwave last December, the artificial intelligence company DeepMind released astounding results from an updated version of the machine-learning chess project today.
The results leave no question, once again, that AlphaZero plays some of the strongest chess in the world.
The updated AlphaZero crushed Stockfish 8 in a new 1,000-game match, scoring +155 -6 =839. (See below for three sample games from this match with analysis by Stockfish 10 and video analysis by GM Robert Hess.)
AlphaZero also bested Stockfish in a series of time-odds matches, soundly beating the traditional engine even at time odds of 10 to one.
In additional matches, the new AlphaZero beat the"latest development version" of Stockfish, with virtually identical results as the match vs Stockfish 8, according to DeepMind. The pre-release copy of journal article, which is dated Dec. 7, 2018, does not specify the exact development version used.
[Update: Today's release of the full journal article specifies that the match was against the latest development version of Stockfish as of Jan. 13, 2018, which was Stockfish 9.]
The machine-learning engine also won all matches against "a variant of Stockfish that uses a strong opening book," according to DeepMind. Adding the opening book did seem to help Stockfish, which finally won a substantial number of games when AlphaZero was Black—but not enough to win the match.
AlphaZero's results (wins green, losses red) vs the latest Stockfish and vs Stockfish with a strong opening book. Image by DeepMind via Science.
The results will be published in an upcoming article by DeepMind researchers in the journal Scienceand were provided to selected chess media by DeepMind, which is based in London and owned by Alphabet, the parent company of Google.
The 1,000-game match was played in early 2018. In the match, both AlphaZero and Stockfish were given three hours each game plus a 15-second increment per move. This time control would seem to make obsolete one of the biggest arguments against the impact of last year's match, namely that the 2017 time control of one minute per move played to Stockfish's disadvantage.
With three hours plus the 15-second increment, no such argument can be made, as that is an enormous amount of playing time for any computer engine. In the time odds games, AlphaZero was dominant up to 10-to-1 odds. Stockfish only began to outscore AlphaZero when the odds reached 30-to-1.
AlphaZero's results (wins green, losses red) vs Stockfish 8 in time odds matches. Image by DeepMind via Science.
AlphaZero's results in the time odds matches suggest it is not only much stronger than any traditional chess engine, but that it also uses a much more efficient search for moves. According to DeepMind, AlphaZero uses a Monte Carlo tree search, and examines about 60,000 positions per second, compared to 60 million for Stockfish.
An illustration of how AlphaZero searches for chess moves. Image by DeepMind via Science.
What can computer chess fans conclude after reading these results? AlphaZero has solidified its status as one of the elite chess players in the world. But the results are even more intriguing if you're following the ability of artificial intelligence to master general gameplay.
According to the journal article, the updated AlphaZero algorithm is identical in three challenging games: chess, shogi, and go. This version of AlphaZero was able to beat the top computer players of all three games after just a few hours of self-training, starting from just the basic rules of the games.
The updated AlphaZero results come exactly one year to the day since DeepMind unveiled the first, historic AlphaZero results in a surprise match vs Stockfish that changed chess forever.
Since then, an open-source project called Lc0 has attempted to replicate the success of AlphaZero, and the project has fascinated chess fans. Lc0 now competes along with the champion Stockfish and the rest of the world's top engines in the ongoing Chess.com Computer Chess Championship.
CCC fans will be pleased to see that some of the new AlphaZero games include "fawn pawns," the CCC-chat nickname for lone advanced pawns that cramp an opponent's position. Perhaps the establishment of these pawns is a critical winning strategy, as it seems AlphaZero and Lc0 have independently learned it.
DeepMind released 20 sample games chosen by GM Matthew Sadler from the 1,000 game match. Chess.com has selected three of these games with deep analysis by Stockfish 10 and video analysis by GM Robert Hess. You can download the 20 sample games at the bottom of this article, analyzed by Stockfish 10, and four sample games analyzed by Lc0.
Update: After this article was published, DeepMind released 210 sample games that you can download here.
We are also releasing 210 new chess games - including a top 20 selected by GM Matthew Sadler @gmmds - that show off its dynamic playing style and we hope will inspire chess players of all levels around the world. https://t.co/ZJDoaon5z0
— DeepMind (@DeepMindAI) December 6, 2018
Selected game 1 with analysis by Stockfish 10:
Game 1 video analysis by GM Robert Hess:
Selected game 2with analysis by Stockfish 10:
Game 2 video analysis by GM Robert Hess:
Selected game 3 with analysis by Stockfish 10:
Game 3 video analysis by GM Robert Hess:
IM Anna Rudolf also made a video analysis of one of the sample games, calling it "AlphaZero's brilliancy."
The new version of AlphaZero trained itself to play chess starting just from the rules of the game, using machine-learning techniques to continually update its neural networks. According to DeepMind, 5,000 TPUs (Google's tensor processing unit, an application-specific integrated circuit for article intelligence) were used to generate the first set of self-play games, and then 16 TPUs were used to train the neural networks.
The total training time in chess was nine hours from scratch. According to DeepMind, it took the new AlphaZero just four hours of training to surpass Stockfish; by nine hours it was far ahead of the world-champion engine.
For the games themselves, Stockfish used 44 CPU (central processing unit) cores and AlphaZero used a single machine with four TPUs and 44 CPU cores. Stockfish had a hash size of 32GB and used syzygy endgame tablebases.
AlphaZero's results vs. Stockfish in the most popular human openings. In the left bar, AlphaZero plays White; in the right bar, AlphaZero is Black. Image by DeepMind via Science. Click on the image for a larger version.
The sample games released were deemed impressive by chess professionals who were given preview access to them. GM Robert Hess categorized the games as "immensely complicated."
DeepMind itself noted the unique style of its creation in the journal article:
"In several games, AlphaZero sacrificed pieces for long-term strategic advantage, suggesting that it has a more fluid, context-dependent positional evaluation than the rule-based evaluations used by previous chess programs," the DeepMind researchers said.
The AI company also emphasized the importance of using the same AlphaZero version in three different games, touting it as a breakthrough in overall game-playing intelligence:
"These results bring us a step closer to fulfilling a longstanding ambition of artificial intelligence: a general game-playing system that can learn to master any game," the DeepMind researchers said.
You can download the 20 sample games provided by DeepMind and analyzed by Chess.com using Stockfish 10 on a powerful computer. The first set of games contains 10 games with no opening book, and the second set contains games with openings from the 2016 TCEC (Top Chess Engine Championship).
PGN downloads:
20 games with analysis by Stockfish 10:
4 selected games with analysis by Lc0:
Love AlphaZero? You can watch the machine-learning chess project it inspired, Lc0, in the ongoing Computer Chess Championship now.
FAQs
After four hours of training, DeepMind estimated AlphaZero was playing chess at a higher Elo rating than Stockfish 8; after nine hours of training, the algorithm defeated Stockfish 8 in a time-controlled 100-game tournament (28 wins, 0 losses, and 72 draws).
Is Stockfish better than AlphaZero now? ›
Stockfish 12 is better than Alpha Zero and would have a higher Elo. You can prove this very easily. Stockfish 12 wins much more games against Stockfish 8 (the opponent of Alpha Zero) then Alpha Zero had won against Stockfish 8.
Is Stockfish 12 better than AlphaZero? ›
Stockfish 12 would be the clear winner. Stockfish 12 beats Stockfish 8 (which was AlphaZero's opponent) more dominant than AlphaZero had beaten Stockfish 8.
What is AlphaZero Elo rating? ›
Game of Shogi on ELO Ratings
Rank | Model | ELO Rating |
---|
1 | AlphaZero | 4650 |
What Elo is Stockfish 14? ›
According to Computer Chess Rating List, Stockfish 14 has an Elo rating of 3549 so an additional 17 points put Stockfish 14.1 at around 3566 – that's a huge 684 point lead over the best human chess player, Magnus Carlsen.
What Elo is Stockfish 15? ›
Ratings were: Stockfish 13 was clocked at elo 3546, Stockfish 14 at elo 3549 and Stockfish 15 at a whopping elo of 3585!!!
What is the strongest chess AI? ›
AlphaZero, the game-playing AI created by Google sibling DeepMind, has beaten the world's best chess-playing computer program, having taught itself how to play in under four hours.
What is the Elo of Stockfish 8? ›
Stockfish 8's elo rating on computer chess rating lists is about 3378, giving AlphaZero a rating of about 3430.
Is Stockfish a brute force? ›
Well, now the raw power of the traditional brute-force Stockfish has been improved by the evaluation abilities of a neural network engine—a mind-boggling combination! Stockfish 12 has added neural network evaluation abilities.
Can a human beat AlphaZero? ›
Since the strongest human on the planet, Magnus Carlsen, doesn't even try to beat Alpha Zero, it's a pretty safe bet that no human will ever again beat these computers on their strongest level.
AlphaZero is the subject of much controversy in the chess world. Some today (a majority of our members!) insist that it is still the strongest chess engine the world has ever seen, that Google DeepMind's chess-playing neural network is still superior to the latest versions of Stockfish and Leela Chess Zero.
Can you beat Stockfish 13? ›
no, human players can't be able to beat it using classical match timing,they even can't draw. But Komodo chess beaten stockfish many times.
Is AlphaZero better than Stockfish 14? ›
AlphaZero also bested Stockfish in a series of time-odds matches, soundly beating the traditional engine even at time odds of 10 to one. In additional matches, the new AlphaZero beat the "latest development version" of Stockfish, with virtually identical results as the match vs Stockfish 8, according to DeepMind.
What Elo was Deep Blue? ›
The Elo rating of chess players, a score used to calculate the skill of chess players, has been creeping up he explained. At his peak, Kasparov was about 2,851 by the turn of the century. Magnus Carlsen, the current world champion had a peak Elo score of 2882 when rated in 2014. Deep Blue's score was above 2700.
How strong is AlphaZero? ›
In each evaluation, AlphaZero convincingly beat its opponent: In chess, AlphaZero defeated the 2016 TCEC (Season 9) world champion Stockfish, winning 155 games and losing just six games out of 1,000. To verify the robustness of AlphaZero, we also played a series of matches that started from common human openings.
How strong is Deep Fritz 14? ›
Approximately 100 Elo points stronger than previous versions. Think Magnus Carlsen's Elo rating of 2870 is impressive? Deep Fritz 14's rating is an astonishing 3150 Elo. Access to the 'Let's Check' (with 200 million extensively analysed positions) and ChessBase engine Cloud.
What is Stockfish's ELO? ›
Stockfish 13 – Elo 3546
Stockfish is the strongest free chess engine. It is easy to say that Stockfish 13 is well beyond the skill of any grandmaster.
Is there a Stockfish 15? ›
Big news for Stockfish fans! Stockfish 15 has been launched today and is already available on Chessify. You may analyze with the new Stockfish version at up to 1,000,000 kN/s speed on chessify.me/analysis. According to the official website, Stockfish 15 is 36 Elo points ahead of Stockfish 14.
Who defeated Stockfish? ›
In the third edition, Stockfish made it to the finals, but was defeated by Leela Chess Zero after blundering in a 7-man endgame tablebase draw.
What rating is Stockfish Level 4? ›
I think Stockfish level 4 - announced with a rating of 1700 - is stronger than that.
This makes Stockfish the strongest chess engine running on your smartphone or normal desktop PC, and we estimate that on a modern four cores CPU, Stockfish 11 could give 1:1000 time odds to the human chess champion having classical time control, and be on par with him.
Is chess AI unbeatable? ›
Chess computers are now so strong that they are practically unbeatable. It is highly unlikely that even the greatest human players would beat a computer playing at a full capacity. This is because a computer can analyze millions of possibilities and compare them against each other within seconds.
What rating is Stockfish level 6? ›
What is the rating of this level? You can see the rating when you look into your played games. Stockfish 6 is rated at 1900.
Can a human beat AI in chess? ›
No human has beaten a computer in a chess tournament in 15 years. In new research, a team including Jon Kleinberg, the Tisch University Professor of Computer Science, developed an artificially intelligent chess engine that doesn't necessarily seek to beat humans – it's trained to play like a human.
What is the Elo of Stockfish 1? ›
The best chess engines for Under 1250 elo rating player is Stockfish 1.01, with an elo of 2785!!! Some people will ask: "Why not use Stockfish 1.0?
Can a human beat Stockfish level 8? ›
Stockfish 8 CANNOT be beaten by a human in such fast time controls. Even strong GMs will face immense difficulty. It is definitely not impossible but it is highly improbable.
What rating is Stockfish Level 5? ›
Level 5: Level 5 is for intermediate players, but it's weaker than level 4 in some ways. Level 6: It's like if levels 5 and 4 put their heads together, and teamed up against you.
Is there a chess engine better than Stockfish? ›
AlphaZero shocked the world after easily defeating Stockfish—the strongest engine at the time—in a 100-game match.
How does chess com detect cheating? ›
Chess.com's fair-play system is thorough, complex and rigorously verified by more than eight years of data from millions of games played by our own members online. Our system gathers and reviews different types of data and other information pulled automatically (and manually) from all member games.
Does Stockfish use Alpha-Beta pruning? ›
Stockfish uses alpha-beta pruning to reduce the number of states it has to look at. Both players will maintain two values alpha and beta representing the minimum value the player looking to maximize the value is guaranteed and the maximum value the player looking to minimize is guaranteed respectively.
AlphaZero would go on to defeat Stockfish in a second match consisting of 1,000 games; the results were published in a paper in late 2018. The AlphaZero-Stockfish matches changed the chess world. Unfortunately, AlphaZero is not available to the public in any form.
Can AlphaZero solve chess? ›
No, it is impossible, even for Alpha Zero. Solving chess in a scientific sense means to prove the value of every possible move in every possible position.
How many neurons are in AlphaZero? ›
Based on the data published for AlphaGo Zero (AlphaZero's Go-playing predecessor) AlphaZero's neural network probably has up to 80 layers, and hundreds of thousands of neurons. Do the math and realize that this means hundreds of millions of weights.
How many games AlphaZero played? ›
At first it made random moves. Then it started learning through self-play. Over the course of nine hours, the chess version of the program played forty-four million games against itself on a massive cluster of specialized Google hardware.
Can you buy AlphaZero? ›
AlphaZero is no product that will be sold. It's on the chess community to develop an AlphaZero-like engine.
Why is AlphaZero not public? ›
Alphazero is not available online, and we are not talking the open source code that could have been available on Github. NO, it is not available at all (it's basically proprietary )!
Did a human ever beat Stockfish? ›
But here we have now a human being able to win against Stockfish 4: Lyudmil Tsvetkov commented 15 winning games played against the monster Stockfish 4.
What is the best chess engine 2022? ›
In this article, we analyze the top 10 strongest chess engines in 2022.
...
- Xiphose. Xiphose has a CCRL rating of 3324 and CEGT rating of 3193. ...
- Shredder. Developed in 1993, Shredder is a chess engine with a CCRL rating of 3324 and a CEGT rating of 3153. ...
- Boot. ...
- Rybka. ...
- Stockfish. ...
- Andscacs. ...
- Fire. ...
- Komodo.
Can humans ever beat Stockfish? ›
The only possible ways a titled player could hope to defeat an assumed strongest chess engine (like Stockfish) would be if they had a stronger engine or if they had some sufficient compensation (at least two pawns).
Is AlphaZero still the best chess engine? ›
AlphaZero is the subject of much controversy in the chess world. Some today (a majority of our members!) insist that it is still the strongest chess engine the world has ever seen, that Google DeepMind's chess-playing neural network is still superior to the latest versions of Stockfish and Leela Chess Zero.
AlphaZero, the game-playing AI created by Google sibling DeepMind, has beaten the world's best chess-playing computer program, having taught itself how to play in under four hours.
Is Leela better than Stockfish? ›
As of January 2022, Leela Chess Zero has played over 500 million games against itself, playing around 1 million games every day, and is capable of play at a level that is comparable with Stockfish, the leading conventional chess program.
...
Leela Chess Zero.
Original author(s) | Gian-Carlo Pascutto, Gary Linscott |
---|
Website | lczero.org |
11 more rows
Can we play against AlphaZero? ›
No, you can't play against AlphaZero, unless you get a special access from DeepMind. Show activity on this post. According to this website, AlphaZero was one of the strongest chess engine and defeated Stockfish (28 victories, 72 draws, and no losses). It runs on Google's supercomputer and is developed by DeepMind.
What is the Elo of Stockfish 8? ›
Stockfish 8's elo rating on computer chess rating lists is about 3378, giving AlphaZero a rating of about 3430.
What Elo rating is Stockfish? ›
1. Stockfish 13 – Elo 3546. Stockfish is the strongest free chess engine.
Is Stockfish a brute force? ›
Well, now the raw power of the traditional brute-force Stockfish has been improved by the evaluation abilities of a neural network engine—a mind-boggling combination! Stockfish 12 has added neural network evaluation abilities.
Has anyone defeated Stockfish? ›
no, human players can't be able to beat it using classical match timing,they even can't draw. But Komodo chess beaten stockfish many times.
What rating is Stockfish Level 4? ›
I think Stockfish level 4 - announced with a rating of 1700 - is stronger than that.
Is AlphaZero better than Stockfish 14? ›
AlphaZero also bested Stockfish in a series of time-odds matches, soundly beating the traditional engine even at time odds of 10 to one. In additional matches, the new AlphaZero beat the "latest development version" of Stockfish, with virtually identical results as the match vs Stockfish 8, according to DeepMind.
Is Stockfish stronger than Lc0? ›
Leela Chess Zero defeated Stockfish with a score of 106-94 in the final of the 13th Chess.com Computer Chess Championship that finished earlier this week. The match, played on the Chess.com server, saw 26 decisive games.
The lowest possible rating is 100. The highest possible rating (in theory) is 3000, although the highest rating any chess player has managed to achieve was 2851 which was held by the World Champion at the time, Garry Kasparov.
Why is Stockfish called Stockfish? ›
The word stockfish is a loan word from West Frisian stokfisk (stick fish), possibly referring to the wooden racks on which stockfish are traditionally dried or because the dried fish resembles a stick.
Why is AlphaZero not public? ›
Alphazero is not available online, and we are not talking the open source code that could have been available on Github. NO, it is not available at all (it's basically proprietary )!
How many games AlphaZero played? ›
At first it made random moves. Then it started learning through self-play. Over the course of nine hours, the chess version of the program played forty-four million games against itself on a massive cluster of specialized Google hardware.
How many neurons are in AlphaZero? ›
Based on the data published for AlphaGo Zero (AlphaZero's Go-playing predecessor) AlphaZero's neural network probably has up to 80 layers, and hundreds of thousands of neurons. Do the math and realize that this means hundreds of millions of weights.