14 December 2018

DeepMind Videos

Last week's post, AlphaZero Is Back!, ended with a request for more time to understand what had just happened.

This is too much new material to digest in the time available for a simple blog post, so I'll come back to the subject as soon as I can.

This video from Google's DeepMind is partly a restatement of what we learned from their first announcement a year ago, partly a statement of what they have been doing since then, and partly a declaration about where they want to go with the technology.


AlphaZero: Shedding new light on the grand games of chess, shogi and Go (4:38) • 'Published on Dec 6, 2018'

The description explains,

DeepMind's AlphaZero is the successor of AlphaGo, the first computer program to beat a world champion at the ancient game of Go. It taught itself from scratch how to master the games of chess, shogi and Go, beating a world-champion program in each case and discovering new and creative playing strategies that hint at the potential of these systems to tackle other complex problems.

A DeepMind blog post, AlphaZero: Shedding new light on the grand games of chess, shogi and Go (deepmind.com/blog), bearing the same title and publication date as the video, goes into more depth. One paragraph explains the essence of the technology.

An untrained neural network plays millions of games against itself via a process of trial and error called reinforcement learning. At first, it plays completely randomly, but over time the system learns from wins, losses, and draws to adjust the parameters of the neural network, making it more likely to choose advantageous moves in the future.

In other words, an NN plays a few million games, compares its predictions about the outcome of its moves against the result of those games, adjusts its internal NN parameters to eliminate discrepancies between its predictions and its results, then starts the process over with the new parameters. Eventually it reaches a level where the predictions and the results almost coincide. DeepMind has also put together a couple of video courses on the underlying technology:-

I now know what I'll be doing during the year-end holidays.

13 December 2018

Endgame Studies with Timman

GM Jan Timman is best known as a player. Last year, in 1967 World Juniors (October 2017), I gave a brief overview of his career as a World Championship candidate, stopping just before his ill-fated participation in the 1993 Karpov - Timman FIDE Title Match.

After being a player, GM Timman is also well known as an author. His Wikipedia page, Jan Timman, currently lists nine book titles and I am certain that the list is not complete. The last book listed is 'The Art of the Endgame', subtitled 'My Journeys in the Magical World of Endgame Studies' (New in Chess, 2011). The book brings us to a third aspect of his dedication to chess : as a composer of endgame studies. Chapter 1 of the 2011 book, titled 'Miniature Studies', starts,

A miniature study is a study with no more than seven pieces in the starting position. With minimal material, the composer must weave the maximum amount of finesses into the position. A classical example is the following study by the brilliant Russian composer Mark Liburkin.

Any study with seven pieces or less is solved by current tablebase technology, which we saw a few months ago in Seven-piece Tablebase on Lichess (August 2018). The Liburkin study is the first position in the first chapter of Timman's book.


Liburkin, '64' 1933
White to play and win

I'll continue to explore 'The Art of the Endgame' and will report any findings worth further research. I've already discovered that seven-piece tablebase positions are paricularly fruitful for further investigation.

11 December 2018

Null Moves

In yesterday's post, Kasparov vs. the Early Engines, one of the encounters mentioned by GM Kasparov in his book Deep Thinking was:-

1992-12 Match vs. Fritz 2, Cologne; +26-11=3 (?)

Of the two mentions of the match in the book, the first (loc.1880 using the Kindle location attribute) starts like this:-

In 1992, I played a long casual blitz match against one of this new generation of PC programs, one that would go on to become nearly synonymous with PC chess engines. Fritz was published by ChessBase, which explains the sardonic German nickname. Its creator was a Dutchman, Frans Morsch, who had also written programs for tabletop chess machines like Mephisto. As such, he was used to having to cram tightly optimized code into very limited resources. He also helped pioneer several of the search enhancements that allowed chess machines to keep improving despite the increasing branching factor that was supposed to slow them down.

This introduction to Fritz and its search heuristics continues with a discussion of the 'null move' concept:-

One of these is worth a brief technical detour because it's an interesting example of how machine intelligence has been augmented in ways that have nothing to do with the workings of the human mind. Called the "null move" technique, it tells the engine to "pass" for one side. That is, to evaluate a position as if one player could make two moves in a row. If the position has not improved even after moving twice, then it can be assumed that the first move is a dud and can be quickly discarded from the search tree, reducing its size and making the search more efficient. Null moves were used in some of the earliest chess programs, including the Soviet Kaissa. It's elegant and a little ironic that algorithms designed on the principle of exhaustive search are augmented by being less exhaustive.

Humans use a very different heuristic when making plans. Strategic thinking requires setting long-term goals and establishing milestones along the way, leaving aside for the moment how your opponent, or business or political rivals, might respond. I can look at a position and think, "Wouldn't it be great if I could get my bishop over there, my pawn up there, and then work my queen around to join the attack." There are no calculations involved yet, only a type of strategic wish list. Only then do I begin to work out whether it's actually possible and what my opponent might do to counter it.

The first point that caught my attention here was the assertion that a null move has 'nothing to do with the workings of the human mind'. When I'm using an engine to evaluate a position, I often inject a null move to identify the opponent's threats, like threatening mate in four or threatening a Knight fork. It's a useful technique that simulates a mental process that occurs after nearly every move in a game between humans: 'What's the threat?'

The following diagram illustrates the null move at the earliest stage of the game, the start position. It shows the position arising after two possible sequences, both of which use the null move (represented by '--').


1.e4 -- 2.d4 // 1.d4 -- 2.e4

As an added bonus, the diagram shows White's main threat after both 1.e4 and 1.d4, which is to advance the other center Pawn and make a strong central position with plenty of space behind the Pawns to develop the other pieces. After either of those moves, Black's objective is to prevent White from achieving that strong center unhindered. This is the underlying idea behind many of the most common opening variations after either 1.e4 or 1.d4.

As for the last paragraph I quoted from Kasparov's book, it starts 'Humans use a very different heuristic when making plans'. In fact, this type of strategic thinking is also possible using an engine by injecting a series of null moves, thereby preventing the opponent from making any moves at all. Here's an example, again using an engine on the traditional start position:-

1.e4 -- 2.d4 -- 3.Nf3 -- 4.Bd3 -- 5.O-O -- 6.c4 -- 7.Nc3 -- 8.Re1

After that sequence of opening moves, the engine I was using gave White's position a value of +3.00. In other words, White's advantage after eight straight developing moves is nearly the same as the value of a minor piece. With a series of null moves, the engine is helping the human to answer the question, 'What's the plan?'

One concept that applies mainly to engines is called 'Null Move Pruning', which I once covered in Chess Engines : Pruning (September 2015). Even here, the concept is similar to what humans do when they avoid looking at a move because it doesn't address the main threat.

10 December 2018

Kasparov vs. the Early Engines

Last week's post, Defending the Human Race?, about Garry Kasparov's two matches against IBM's Deep Blue computer, reminded me that I had an open follow-up from an earlier post this year: Kasparov vs. Hsu (February 2018). That post compared milestones in the evolution of Deep Blue that are found in both Kasparov's book Deep Thinking and Feng-hsiung Hsu's book Behind Deep Blue. The post closed by saying,

Kasparov's book also gives details about his games/matches against other chess computers. I should compare this to my page Garry Kasparov's TMER.

That cryptic acronym 'TMER' stands for Kasparov's Tournament, Match, and Exhibition Record (1973-; Last updated 2014-08-11), a record of Kasparov's career that I've been maintaining on-and-off since the year 2000. The following chart merges references in Kasparov's book with the corresponding data in the TMER.

The 'LOC' references, like the first one '>>> LOC0038', refer to locations in the Kindle version of the book, which is the version I've been working from. I suppose they can be translated to page numbers in the hardcopy version of the book, but I don't know how to do that easily. For explanations of the other codes in the chart, see the TMER page.

The chart shows that Kasparov's book mentions six events where he played against computers before the two famous matches against Deep Blue. If I find any more references in the book, I'll update this current post. In any case, I'll come back to the chart in another post.

09 December 2018

Puzzle Rush

Our featured November video on this blog was 'The World Is Watching', about the start of game one of the 2018 Carlsen - Caruana match for the World Championship. For the December video, let's skip ahead to the match tiebreak.


Puzzle Rush #1: World Chess Championship edition! (7:47) • 'Published on Nov 27, 2018'

The video from the John Bartholomew channel starts,

This is John! I'm back from London. The tiebreaker for the 2018 World Chess Championship is tomorrow, Wednesday, November 28. I'm very much looking forward to it and I'm sure you are as well. What does that have to do with Puzzle Rush? It's been sweeping the chess community and is a new feature on Chess.com where you try to solve as many puzzles as possible in a five minute spin.

The answer to the question 'What does the World Championship have to do with Puzzle Rush?' lies in the following tweet.

While that idea has as much chance of being realized as having all regulation games in the next WCC match end decisively, we can still dream. For more about the new speed game, see Puzzle Rush - Compete to solve Chess Puzzles (chess.com).

07 December 2018

AlphaZero Is Back!

Ding, ding, ding! As I made my way through this morning's reading list on chess topics, the bells were sounding everywhere. Almost a year to the day after their first shock announcement, Google's Deepmind had just released more news about AlphaZero, by all reports the strongest chess player ever.

The news was propagated via Science magazine -- see Table of Contents : December 07, 2018 (sciencemag.org; cover shown on the left) -- which included three articles by world class authorities on computer chess:-

The description for the cover of Science said,

Starting from random play and given no domain knowledge except the game rules, the AlphaZero program taught itself to play chess, shogi, and Go, defeating a world champion program in each game. Blue translucent pieces represent AlphaZero's possible moves; percentages indicate the predicted outcome. A single algorithm that can master several complex problems is an important step toward creating a general-purpose machine learning system to tackle real-world problems. • Image: DeepMind Technologies Limited

This is too much new material to digest in the time available for a simple blog post, so I'll come back to the subject as soon as I can.

06 December 2018

Breaking the 2800 Barrier

A few months ago I opened Breaking the 2700 Barrier (June 2018), by saying,

No, I'm not talking about achieving a 2700 rating. I'm talking about post no.2700 on this blog.

I continued by discussing the history of chess players rated 2700 or more, then closed the post saying,

To break 2800, all I have to do is write another 100 posts.

Post no.2800 was A Conversation with Demis Hassabis a few weeks ago, but given the flurry of posts on this blog for the 2018 Carlsen - Caruana match that finished last week, I'm only finding spare time now. In a nutshell, the following chart shows the history of players rated 2800 or more.

The first column shows the evolution of top ratings during the second half of the 1980s, taken from the January list for each year. Throughout that period there were only two players rated over 2700, and Garry Kasparov was the first to break the 2800 barrier at the end of that decade.

The second column shows the top players after further intervals of five years: 1995, 2000, and 2005. Kasparov continued to head the list, but the number of 2700 players expanded steadily.

A little 'i' after a player's name means 'inactive'. FIDE hasn't always been consistent with the 'i' flag and Bobby Fischer made a sudden appearance in 2005, perhaps because he was in the news for having been detained in Japan.

A little 'w' means 'woman'. It appears only once on the chart, in 2005, when Judit Polgar made the top-10 list of all players. Note that there are not separate rating systems for men and women. All tournaments, even when restricted to women (where Judit Polgar never participated), are rated using the same methods and criteria.

The last column shows top-10 lists for the current decade: 2010, 2015, and 2018. For the first time we see players other than Kasparov rated over 2800.

I'll be back in another 100 posts to write 'Breaking the 2900 Barrier'. It promises to be a short post.

04 December 2018

December 1968 'On the Cover'

Here we are 'wrapping up' another year of covers from American chess magazines 50 years ago. Last year, in December 1967 'On the Cover' (OTC; CR featuring 'Manhattan Chess Club President Jacques Coe'), we were spared the funky green cover seen in December 1966 OTC -- not to mention previous Decembers -- but this year the tradition returns.


Left: 'Bent Larsen Interviewed' (Drawing by Bob Brandreth)
Right: 'Merry Christmas ... to the World of Chess'

Chess Life

The year 1968 was the year of Bent Larsen. Two months ago, in October 1968 OTC, he was featured on both covers, and this month he returns. I've already excerpted a big chunk of the interview ('The Larsen Opinion : An Interview with Bent Larsen' by Ben Crane) in a post on my World Chess Championship blog, An Organization of Amateurs (April 2017). Here I'll add the introduction.

Crane: You have recently become a "chess professional." Most masters, however, have found that they need another line of work to supplement their income. What factors led you to this decision?

Larsen: I don't know whether you can say I've become a chess professional recently. I have made a living as a chess player since I left high school when I was seventeen. But it was only about four or five years ago that it became a real profession, because before that I was just living as a modest student. It has not been a decision that I have taken suddenly, I just drifted into it. Now it appears that is very possible in Denmark to make a living from chess playing, chess writing, and sometimes on radio and TV. Of course it is better when you are winning money prizes than when you don't do well, but I think theoretically it would be possible for me to make a living as a chess player without playing at all -- just writing, giving exhibitions, and so on.

Chess Review

After the green cover, there was little inside CR about Christmas. The 'Announce the Mate!' puzzle page had a Christmas-y title to a non-Christmas-y poem.

MERRY CHRISTMATES!
The outlook is not brilliant for your chessic foes today.
The score stands 0-0, but they really have no play.
So on the gloomy cross-board foe glum melancholy sits
As it seems but little like you'll miss on any hits.
Your move precedes their turn and calls on you to mate.
So smite them hip and thigh to send 'em to their fate.
And rack up a good score on the old sockdolager slate.

Poems of a similar style introduced each of the ten puzzles on the page.

03 December 2018

'Defending the Human Race?'

In the latest post on 'Top eBay Chess Items by Price', 1958 Portoroz Interzonal, I had a second item on the short list. Titled 'Extremely rare original program for Kasparov vs. Deep Blue Computer Chess Match', it sold for around US $375, 'Best offer accepted'. The description said,

This is an extremely rare original brochure from the match, which was held on May 3–11, 1997 in New York City. This match will echo forever in the history of chess, and mankind, as the first defeat of a reigning world chess champion by a computer under tournament conditions. Several books have already been written of this historic event, including one by Kasparov himself. The audience was small (I was there) and probably not many brochures were kept.

Sensing the historic importance of the event, I made a point to keep mine in pristine condition, but I have never seen another -- other than a copy in a museum -- the Computer Museum. In like new condition, with white glossy pages. 24 pages. The brochure measures 8 1/2 x 11". A special item for collectors of chess or computer and AI memorabilia

The description pointed to a page from the Computer History Museum, Defeating the World Chess Champion | Mastering the Game (computerhistory.org), where the programs for both Kasparov - Deep Blue matches (1996 and 1997) can be downloaded. Here are the front covers of the two programs.


Left: 1996 match; Right: 1997 match

The 'Welcome' page of the 1997 program says,

Greetings! On behalf of IBM. I would like to welcome you to the IBM Kasparov vs. Deep Blue Rematch. Whether you're with us in New York City or participating in the match via the Internet, we're glad you've joined as for this historic event. We're anticipating a very exciting match.

We at IBM are quite proud of Deep Blue and the team of research scientists who have created it. They are among the hundreds of IBM researchers who work everyday on building, inventing and discovering new technologies which improve our lives and help define the future of computing.

So why has IBM built a machine that plays chess? The applications of the technology we are using in this experiment go far beyond a game. The RS/6000 SP system, the technology underneath Deep Blue, is being used everyday to run businesses. perform technical analysis and conduct scientific experiments. By emulating the Deep Blue architecture -- boosting the power of the SP with specialized processors and software knowledge -- we have plans to apply this awesome power to a variety of disciplines, and expect to see amazing results. By introducing the Deep Blue technology to pharmaceutical drug design, financial risk assessment and decision support problems, for example, we hope to realize faster drug development, stronger economic forecasting and smarter decision making. The possibilities are virtually endless.

There has been much discussion in both public and private forums regarding the significance of this match. Are machines taking over mankind? Have we unleashed a monster? Is Garry Kasparov defending the human race?

We at IBM are very pleased that our experiment has caused such a stir. To us, Deep Blue represents innovation leadership and the unlimited promise of technology. It is not simply man vs. machine, but man and machine striving together to reach new heights -- and reaching them. Welcome to the future.

Sincerely,
Paul M. Horn, Senior Vice President, Research

I once recorded the results of the two matches on my World Chess Championship site: Kasparov vs. IBM's Deep Blue. I wrote,

It's undoubtedly a minority opinion, but some informed observers of the international chess scene believe that Kasparov lost his claim to the World Chess Champion title when he lost his second match to IBM's Deep Blue computer.

More than 20 years later, I'm still not sure what to think. We say 'World Women's Chess Champion' and 'World Junior Chess Champion' when talking about restricted titles. Shouldn't we be saying 'World Human Chess Champion' and leave the unrestricted title to the machines?

02 December 2018

1958 Portoroz Interzonal

We often see autographs in this ongoing series on Top eBay Chess Items by Price (March 2010), but I can't remember an item where we saw three autographs for each player in a tournament. Titled 'Extremely rare!! Portorose 1958 tournament full set; autographs all chess players, Portorose Yugoslavia', the auction consisted of a half-dozen items, including a signed program and two signed cards.

The full package listed for US $4000 and sold for something between $2000 and $2600, 'Best offer accepted'. To illustrate the auction, I'm featuring the front of a postcard, because it is more visually interesting than the signatures on the back of the card.

The description said,

This is an extremely rare, probably unique lot of chess related historical memorabilia of museum quality. The lot includes the program of the tournament, a postcard of the tournament with all the players pictured, and the official tournament envelope with all the signatures. These items are all fully signed! Even the judges are signed!

A "tourist info" brochure on Portorose - Yugoslavia and two other leaflets comes with the lot. All this was purchased from the same owner who attended the tournament as a visitor.

I compared the names in the program against my page on the 1958 Portoroz Interzonal Tournament, which lists 21 players. The program had 20 photos of players over their autographs, plus Golombek & Vukovic. Benko was missing completely. All players plus Golombek are pictured on the postcard above. This auction complements a previous post for 'Top eBay Chess Items', 1959 Yugoslavia Candidates (December 2017):-

Titled 'Chess book signed by eight masters, incl. Fischer, Keres, Petrosian, Smyslov and Tal', it sold for around $1500, 'Best offer accepted'.

Tal won the Interzonal, won the Candidates, and went on to snatch the World Champion title from Botvinnik. That's Tal ('Talj') in the upper right corner of the postcard.

30 November 2018

An AI/NN Blockbuster

Last week's AI/NN post, A Conversation with Demis Hassabis, quoted CEO's Deepmind saying, 'We had a couple of very strong chess players come in and look at the [AlphaZero] games and help us analyze them.' This week we learned the likely identity of one of those strong players -- Matthew Sadler -- who applied AlphaZero's analysis to the first eight games of the 2018 Carlsen - Caruana match, which finished two days ago. Here is the longest of the three videos that Sadler produced on the match.


DeepMind's AlphaZero on Carlsen-Caruana Games 1, 3, 5 & 8; Sicilian Defence (36:28) • 'Published on Nov 24, 2018'

The description starts,

Two-time British Chess Champion Matthew Sadler uses DeepMind's AlphaZero to analyse Games 1, 3, 5 & 8 (the Sicilian Defence) of the 2018 World Chess Championship between Magnus Carlsen and Fabiano Caruana.

The two other videos are also on Youtube:-

The three videos copy the remainder of the description:-

Matthew analysed the games together with Women's International Master Natasha Regan, his co-author for an upcoming book on Alpha Zero’s approach to chess: Game Changer. Chess24 is working with Matthew and Natasha to share World Championship game insights from AlphaZero with the chess community. We’re happy to have the opportunity to share this during one of the most exciting points in the year for all chess fans. With thanks to DeepMind for their support.

More about the book ('Expected January 2019') can be found at Game Changer: AlphaZero's Groundbreaking Chess Strategies and the Promise of AI (newinchess.com; 'The story behind the self-learning artificial intelligence system with its stunning chess skills'). The book is almost guaranteed to be a blockbuster of a chess book.

29 November 2018

November Yahoos

Mid-month, just after the start of the 2018 Carlsen - Caruana match, we had World Championship Yahoos. Now the last move of the tiebreak has been played, the match is over, and just like last month's October Yahoos, FIDE chess is center stage.

2018-11-24: Caruana ‘Suffers Successfully’ In Game 11 Of The World Chess Championship (fivethirtyeight.com). I've already mentioned this article on my World Chess Championship blog in 2018 Carlsen - Caruana, the Third Week, so I'll just note its lead paragraph.

With his last chance to command the white pieces in a regulation game in the World Chess Championship, defending champion Magnus Carlsen was unable to drum up any attacking chances. Game 11 -- like the 10 that preceded it -- ended in a draw.

2018-11-28: World Chess Championship title to be decided today (yahoo.com; ABC News Videos), or so says the main headline. The secondary headline from a day later says, Heartbreak for American challenger in world chess final (yahoo.com; Good Morning America), which is the main story.

The match is over, Magnus Carlsen crushed Fabiano Caruana in the tiebreak after not winning a single regulation game, and that is all most people will remember about this strange match. Congratulations to GM Carlsen on his fourth consecutive victory in a World Championship match!

27 November 2018

Twelve Straight Draws!

Start with Nine Straight Draws!, add three more draws, and the world begins to wonder...

...'What is wrong with professional chess?' From Mark Crowther's The Week in Chess 1255 (theweekinchess.com; Monday, 26 November 2018):-

The World Chess Championship goes to a final tie-break after the classical part was drawn 6-6. Two years ago Magnus Carlsen took a short draw with white against Sergey Karjakin to force a tie-break he dominated. This time in the final game 12 with black, Carlsen was looking so strong on the board and the clock that he should have been pressing for a win but instead offered a draw on move 31 that Caruana was probably glad to accept.

Carlsen is 91 ratings points stronger than Caruana in rapid chess, it remains to be seen if this translates to an easy win for him on Wednesday. Fabiano Caruana has his chances and may feel he escaped today. Perhaps his rapid rating doesn't reflect his true strength. There's still all to fight for on Wednesday.

From the world chess federation's official site, FIDE WCCM Game 12 review: Relief and Pragmatism (fide.com; 27 November):-

After 31 moves of game 12, Magnus Carlsen offered a draw which was accepted by Fabiano Caruana. All 12 classical games have now been drawn - a result unprecedented in world championship history - and the players head for a tie-break on Wednesday. [...] Carlsen revealed afterwards that his intention before the game started was to hold the draw and head for the tie-breaks, and therefore he was not in the right mind-set to take any risks and play for the win.

Like any sporting or cultural activity -- pick your classification -- the main function of chess is to provide entertainment. Twelve straight draws ... that's not entertainment. What to do? Two suggestions which have been many times in the past are (1) prohibit draw offers (the so-called Sofia rule), and (2) play the match tiebreaks before the regulation games. After nearly two and a half weeks, the chess event of the year will be decided by a short series of rapid and blitz games. Let's hope it doesn't reach the final Armageddon game.

26 November 2018

Carlsen's PGN 2000-2018

I could continue last week's post, Update on Two World Champions, with news about both champions, Stockfish and Magnus Carlsen. Instead I'll just close the Carlsen portion by noting that I included games for the period 2017-2018 in the PGN file for Magnus Carlsen's Tournament, Match, and Exhibition Record (TMER; 2000-).

The previous version of the file contained 2117 Carlsen games played through 2016. The new version adds 554 games. For more about the additional games, see Carlsen's PGN 2017-18 (October 2018).

25 November 2018

Rex and the Game of Kings

'At first glance he may look like just another aging retiree...' • This video could be the best clip I've ever featured in the series on The Sociology of Chess (November 2016). It's not just about chess; it's about reviving a troubled American city.


Rex Sinquefield, the American Chess Mogul (Full Segment) | Real Sports w/ Bryant Gumbel | HBO (14:12) • 'Published on Nov 6, 2018'

The description said,

Chess aficionado Rex Sinquefield has spent his life trying to make the classic game fresh again in America. With the help of Grandmasters Fabiano Caruana and Maurice Ashley, he hopes to get the next Bobby Fischer.

Who needs corporate sponsors when you have individuals like Rex Sinquefield? If you need help with pronunciation, it's 'Sink-field'.

The last game of the 2018 Carlsen - Caruana match will be played tomorrow. With the score at 5.5 each after 11 games, the Sinquefield vision might be realized.

23 November 2018

A Conversation with Demis Hassabis

Continuing a series of posts on AI/NN, I'll suspend the nuts-and-bolts overview seen in the previous post, GPU Benchmarks, to cover a rare chess-related appearance of AlphaZero's guiding light, Demis Hassabis, who visited the 2018 Carlsen - Caruana venue. Later he tweeted, A real honour to make the ceremonial first move of match 8 of the World Chess Championships today... (twitter.com). After making the first move, he visited the commentary room.


World Chess Championship 2018, Day 8, First moves
(youtube.com; 'Streamed live on Nov 19, 2018')

At around 23:40 into the Youtube clip, Hassabis joined the official commentators for the match. Left to right: Judit Polgar (JP), Anna Rudolf (AR), Demis Hassabis (DH).

AR: We are so glad that Mr. Demis Hassabis has returned to our studio. Welcome back, Demis. The co-founder and CEO of Deepmind as we discussed already and if you were not here yesterday I don't what you were doing because you missed a very insightful conversation with one of the brightest minds in the world. We started out with AlphaGo and AlphaGo Zero and today we will discuss further topics about artificial intelligence and chess as well.

DH: It's great to be here and great to do the first move, because it was really interesting feeling the intensity in the room. It's quite amazing the amount of energy in the room. You go inside and you see they're superconcentrating. It was quite an interesting experience on that side of the glass. It's almost claustrophobic inside. It feels like the room is not big enough to contain the energy of the two players.

AR: Did you expect 1.e4 or did you think Fabiano would tell you something else?

DH: I wasn't sure what he was going to do. I'm pleased to see the Sveshnikov because I played this for Black. It was quite a coincidence it was this match [game?] I got to do the first move. We'll see how this pans out.

AR: We will definitely ask for your expert advice. As mentioned we are going to discuss further topics about artificial intelligence. Yesterday we were so sorry when we had to say goodbye to you. We thought that there are so many other fascinating topics about artificial intelligence that we wanted to have you here for hours. Thank you so much for coming back. The first topic I wanted to discuss was how differently AlphaZero is thinking about the game of chess -- if we compare it to humans and if we compare it to computer engines.

DH: There are two interesting things to say about that. One is how many moves do the chess engines calculate per decision. Human grandmasters maybe look at 100 moves, something like that order [of magnitude], to be able to make a decision. Something like Stockfish and traditional chess engines, they look at 10.000.000 moves before they make a decision. AlphaZero is somewhere in between, so it looks at 10.000 moves before it makes a decision. It's not as efficient as human decision making but it's much more efficient than traditional engines. It looks at a lot less moves because it's better at evaluating positions.

The second thing that's interesting about it is because it doesn't have in-built moves. It doesn't have 'a Queen is nine points, a Rook is five points'. It doesn't know anything about those piece values, so it senses everything in the context of the current position. We speculate that it's much easier for it to make long term sacrifices; for example, because it doesn't have to overcome its in-built programming. Say it's going to sacrifice an exchange. An engine like Stockfish would have to calculate that it's going to get enough in return for that two points difference. Whereas AlphaZero doesn't have that rule in-built so it can just decide the Rook is an asset, the Knight is an asset, and in the current position that Knight is a stronger asset for the opponent than my Rook. It can make the sacrifice even if it can't calculate explicity that it's going to get enough compensation. It can just sort of decide contextually that in these kinds of positions that exchange is worthwhile.

JP: How does he make a difference between Knight and Rook if he doesn't know that the Rook is worth more?

DH: I guess it's learned over the millions of games playing against itself that the Knight will give a better outcome over the whole experience. In the context of this particular position it can make a decision to make the sacrifice. Stockfish can do that, too, but it would need to calculate quite carefully that it's going to get enough in return to overcome this in-built rule that you are losing two points.

JP: Does AlphaZero have intuition?

DH: It is sort of like a very intuitive player. It does it more by feel in effect because it's taking the pattern of the current board and deciding that this is worthwhile. It's not necessarily explicitly calculating out. It's more akin to something like intuition in human terms. Of course, it doesn't know anything about intuition or any of these terms we're using -- it's just a computer -- but it's more like that and it comes out in the style of the play. It really likes sacrifices. It's very positional, I would say.

JP: How many moves ahead does it calculate?

DH: It can calculate quite deep lines if it needs to, but it only looks at 10.000 moves per decision. It's quite a lot compared to a human player, but it's much, much less than you're used to with a chess engine which is millions of moves. In order to compensate for that lesser amount of caculation it has to have better evaluation.

JP: What makes the difference about how deeply it goes into certain lines? We have strategic positional lines and we have tactical. Does it understand that in a tactical line you really have to go all the way?

DH: Not explicitly. Chess engines have this extra calculation. They know that if it's an imbalanced position then they should calculate more : there are special moves for that as well. AlphaZero doesn't have anything explicit about that. If it feels that a certain line is unresolved -- it doesn't quite know what's happening -- it can search in more depth. So there is some implicit way it does that, but not in the explicit way that chess engines do. This is something very interesting for us. We've only just built the system, so the next stage over the next year is to try and reverse engineer it to see how it's making its decisions. At the moment it's more like a black box. It makes these decisions, but I would like to know, for example, what does it rate a Rook and a Knight overall. We don't actually know. It doen't express its evaluation in terms of Pawns, like chess engines do. It expresses it in terms of percentage chance of winning.

At this point, AR discussed the games played during the AlphaZero - Stockfish match. See The Constellation of AlphaZero (December 2017), for the earliest posts on this blog about the match.

AR: AlphaZero not only won the match very convincingly, but it came up with these sacrifices that you mention, not just the positional sacrifices we are used to in human games, which would be a Pawn sacrifice, but it went on sacrificing a Bishop, a Knight, for something very long term, not a calculation that there would be a reward for the sacrifice 10 or 20 moves later. Then there was another move I really loved, a Queen move. I felt like we are learning something from this AI bacause that's neither a human move nor an engine move.

DH: Exactly. There were a lot of examples. I think people were surprised that it was making very unusual moves that were kind of alien. They weren't really the kind of moves a computer engine would do. I hope that's going to give strong chess players new ideas, maybe usher in a new era of creativity, because it's a very interesting style.

We're going to release a lot more games and then people can see even further what this style is. One thing it really favors is mobility. It really likes mobility and optionality for its pieces, and it likes restricting the mobility of the opponent, including using Rooks, especially Rooks, on outposts, very advanced outposts, which is quite unusual for chess.

Here the discussion turned to the match between AlphaGo and Lee Sedol, where the Korean player learned from AlphaGo's play. Could anything be learned about chess from AlphaZero?

DH: We had a couple of very strong chess players come in and look at the games and help us analyze them. One thing they told me, that stuck in my mind, is that it felt to them as if the board was much bigger somehow. I thing people will see that when they see the games. It plays on one wing, creates a few weaknesses, then it switches all of its pieces to another wing and makes more weaknesses. Then it finally goes back and the opponent's position collapses because there are a few too many weaknesses. It's very interesting how it controls that situation. People are going to find these newer games, with an even stronger version of AlphaZero quite fascinating.

AR: I believe it was former World Champion Garry Kasparov who said that IBM's Deep Blue basically caused the end of an era but AlphaZero is the beginning.

DH: I hope so. It was very kind of him to say so. Garry has spent a lot of time thinking about computer chess and he was right in the middle of the biggest moment of all. It's been fascinating talking to him about that. I think what he meant was that he realized that we're building these general systems not only just to learn all types of different games, including chess, but eventually to apply them to real world problems in science and medicine. These AlphaZero techniques are not built specifically for chess -- it just learns it for itself -- these techniques can be used for other complex domains in the real world.

After further discussion about how each version of AlphaZero learns from its previous version and what the Elo rating of the AI engine might be, the conversation turned back to the Caruana - Carlsen game. For another video showing Hassabis, see Kasparov Talks at Google (June 2017).

22 November 2018

Nine Straight Draws!

The world isn't watching...

...In fact, I think the match is pretty interesting, but I'm not the world.

20 November 2018

World Championships in London

How many World Championship events have been held in London? Here's what I found on my World Championship site (see link in the sidebar). From the 19th century:-

After 1883 London we have to skip ahead a full century to find the next events:-

On top of those, London has hosted a few restricted events:-

Taking all of that into consideration, it's appropriate that London was awarded the venue for the 2018 Carlsen - Caruana match.

19 November 2018

Update on Two World Champions

Before I close the series on Catching Up with Engine Competitions (October 2018), there is one more event to document. Last week's post, Stockfish Wins TCEC Season 13, was left unfinished because the Chess.com event was still running. Today the site announced Stockfish Wins Computer Chess Championship Blitz:-

The highest-rated chess engine of all time added another title to its resume this week as Stockfish decisively won the Chess.com Computer Chess Championship 2 [CCC 2]: Blitz Battle. The victory comes six weeks after Stockfish won the first revamped CCC: Rapid Rumble in October.

Trailing Stockfish by a large margin, the other three engine finalists finished in the order Komodo, Leela (Lc0), Houdini. The announcement carried news about the next event:-

CCC 2: Blitz Battle will conclude with some bonus games before Chess.com begins CCC 3: Rapid Redux, an all-new championship event with the 16 top engines in the world and a time control of 30 minutes per game plus five seconds increment per move.

The PGN for all three finished stages of CCC2 is available via the same Chess.com page. With TCEC S14 already underway and CCC3 promised soon, the engine to engine competitions will carry on for the next few months.

***

Continuing with an older unfinished series, the overview of Carlsen's career, I went back to Carlsen's PGN 2017-18 (October 2018), and concatenated the World Champion's record for 2017-18 onto the record for his entire career. The result is available in the usual place, Magnus Carlsen's Tournament, Match, and Exhibition Record (TMER; 2000-). The PGN will be added soon.

18 November 2018

2018 Carlsen - Caruana, the Venue

Was last month's Chess and Cubofuturism (October 2018) the last post in this blog's long running Flickr series? When I signed in today to survey the chess items uploaded since the 'Cubofuturism' post, I received the following message:-

Free accounts will soon be limited to 1,000 photos or videos. Flickr isn’t Flickr without the contributions and participation of our free members, and we remain committed to a vibrant free offering. If you are a free member with more than 1,000 photos or videos, you will have ample time to upgrade to Pro (for 30% off your first year) or download your photos and videos.

On top of that, of the 300+ photos newly available in the last month, only two were interesting enough to add to my favorites list. Of those two, neither was suitable for today's post. If a photo doesn't tell a bigger story, it's just an interesting photo. What to do?


Google image search on 'chess holborn college'

I decided to look at images related to the venue for the 2018 Carlsen - Caruana being played in London. All the images shown above, plus the others that I cropped out, are part of a bigger story. For the original announcement about the choice of venue, see News about 2018 Carlsen - Caruana (August 2018), on my World Chess Championship blog.

16 November 2018

GPU Benchmarks

First I got some idea about GPU Pricing in last week's post. Then I got some idea about the AI engines' operating environment in this week's post, Stockfish Wins TCEC Season 13. Both the CCCC and the TCEC offer public information about the setups they use in support of the AI/NN engines.

Computer Chess Championship (chess.com):-

GPU: 4 x Tesla V100 (64 GB GPU memory)
CPU: Intel Xeon @ 2.70GHz
RAM: 256 GB

TCEC Season 14 (chessdom.com):-

GPU: 1 x 2080 ti + 1 x 2080
CPU: Quad Core i5 2600k
RAM: 16GB DDR3-2133

I've listed only the hardware that allows a comparison of the two setups. The V100s used by CCCC were first offered by Nvidia in 2017; the 2080s used by TCEC were first offered in 2018. How do they compare? Here's a chart from the same company that provided the numbers I used in 'GPU Pricing'.


October 2018: 2080 Ti TensorFlow GPU benchmarks - 2080 Ti vs V100 vs 1080 Ti vs Titan V (lambdalabs.com) • 'The 2080 Ti comes out on top as the best GPU in 2018 for training neural nets.'

Although the 2080s don't offer the same throughput as the V100, a cost/benefit comparison improves in favor of the 2080s when you factor in the price of the GPUs. Note that these numbers are for training the NNs listed to the right of the top chart. The performance of a system running a specific chess engine with its trained weights would be different.

15 November 2018

Chess @ 538.com

At the 2018 World Championship we've had four games, all draws, and the assembled journalists search for any scrap of info that might be considered newsworthy. For me, that means looking at the journalists. I ended the previous post, World Championship Yahoos, with,

It's the second time in a month that I've used FiveThirtyEight.com as a reference. The first was Out with the Old!, about the forced retirement of Kirsan Ilyumzhinov. Maybe I should give that news resource a deeper look.

The chess writer for 538.com is Oliver Roeder ('He holds a Ph.D. in economics from the University of Texas at Austin, where he studied game theory and political competition.'). His reports on games three and four demonstrate that he knows his chess:-

A unique feature of the reports is a graphic overview of the match so far. Let's call it, 'How to summarize a World Championship match in 100.000 pixels or less'.


Match overview through game four (538.com)

We can easily see that Carlsen held a significant advantage in game one and that the success of the Black pieces in all games has been the main story so far. But what about that 'Biggest Blunder' in the headline for game four? Was Woody Harrelson knocking over pieces again? No, that pre-game excitement from the first game was overshadowed by a more sinister incident. The report went on to explain,

The day’s humdrum play was overshadowed by some excitement off the board, though. A chess-world controversy -- or at least what qualifies as one -- erupted. Before the game, the posh Saint Louis Chess Club posted, and quickly deleted, a YouTube video appearing to show aspects of Caruana’s pre-championship preparation sessions.

The report included a still shot of a computer screen where the image could be expanded to make everything on the screen readable. And, yes, it's the sort of information that might prove valuable to the opposition. Some of the less knowledgeable comments on the incident wondered, 'Why the fuss?' In brief, because match preparation is opening preparation, nothing more and nothing less -- how do I press for a quick advantage playing White and how do I avoid a quick knockout playing Black? A couple of old posts on this blog delved into the nuts and bolts:-

The resulting brouhaha convinced one respected chess journalist, GM Ian Rogers of Australia, to resign his job working with the American team: @GMIanRogers: Sadly parting ways with @ChessLifeOnline after a decade... (twitter.com):-

...I declined to accept edits to my round 4 World Ch'p report which would downplay responsibility of editors of the Caruana video, downplay the effect of the video on Caruana's chances, and omit the key image from the video.

On top of that, all of the videos produced by the St.Louis Chess Club disappeared from Youtube. Out of sight, out of mind? Hardly. Someone in St.Louis is guilty of an unprofessional lapse of judgement. That's the person who should resign -- not a journalist doing the job he was paid to do.

13 November 2018

World Championship Yahoos

In last month's wrapup of chess news reported by the mainstream press, October Yahoos, I wrote,

Let's hope that 'November Yahoos' will be filled with stories about next month's 2018 Carlsen - Caruana match.

No need to wait until end-November to survey the results. Under headlines like 'Trump's last-minute cancellation draws criticism' and 'Entire teams penalized after wild college football brawl', chess was the top story.

2018-11-10: The World Chess Championship Is Deadlocked After Game 2 (yahoo.com; FiveThirtyEight)

Heavy rain showers, a gentle breeze and 57 degrees [F = 14 degrees C] in London, the BBC reported this morning. The top American grandmaster Fabiano Caruana was unprepared for such weather, arriving for Game 2 of the World Chess Championship sporting a wet blazer. Never mind the sartorial dampness, however. He arrived excellently prepared for the chess.

Along with that main story, the accompanying stories were also about games one and two of the match.

2018-11-10: Chess stars draw again in world tournament (yahoo.com; AFP)

The second round of the 2018 World Chess Championship in London between three-time defending champion Magnus Carlsen and American Fabiano Caruana ended in a draw on Saturday. The two young chess stars, both under 30, have one point each after drawing their opening two games of the tournament

Almost all Yahoo stories have a comments section, and although the comments are often a mindboggling display of ignorance, sometimes they are on target. The story above continued, 'Italian prodigy Carlsen is seeking to cement his reputation as one of history's greatest chess players', and drew snorts of derision like 'Italian, Norwegian they're all the same'.

2018-11-09: World Chess Championship starts with Harrelson blunder (yahoo.com; Reuters)

World Chess Championship officials might rethink the role of ceremonial starter after Hollywood actor Woody Harrelson knocked over a king and moved the wrong pawn in a comical start to the 2018 event in London on Friday

That story gathered more than double the number of comments of the other two stories combined, stuff like:-

  • 'Why rethink? Woody was hired to alter the image a bit, and he did that. Also made the internet news, giving the competition a wider audience, if only briefly!'
  • 'The errors seems a little too convenient. He technically conceded the world championship on the first move. It's just a fun publicity stunt like announcing the wrong Miss Universe.'
  • 'Never knew Woody was a chess man.'

Looks like it's once again time to point out Woody Harrelson vs Garry Kasparov; "Cheers!"; Consultation (1999), Prague CZE (chessgames.com). The game started 1.e4 e5 2.Qh5, the same opening seen in another celebrity game, Boris Becker vs Garry Kasparov; CNN exhibition (2000); New York, NY (ditto).

Back to that first story, it's the second time in a month that I've used FiveThirtyEight.com as a reference. The first was Out with the Old!, about the forced retirement of Kirsan Ilyumzhinov. Maybe I should give that news resource a deeper look.

12 November 2018

Stockfish Wins TCEC Season 13 (and Everything Else)

In Catching Up with Engine Competitions (October 2018), I took a checkpoint on the progress of two major computer chess tournaments. What has happened in the four intervening weeks?

TCEC S13

Stockfish beat Komodo +16-6=78 to win the TCEC Season 13 Superfinal. Only one opening resulted in a win for both engines, an Old Indian (1.d4 d6 2.Nf3 Bg4), played in games 85 and 86. For a report on the superfinal of the previous season, see Stockfish Wins TCEC Season 12 (July 2018).

In the TCEC Cup that was played before the S13 Superfinal started, Stockfish beat Houdini +1-0=7 in the final match of the knockout event. The two engines beat LC0 (Leela) and Komodo in the semifinal matches.

Details for both the Superfinal and the Cup, including PGN game scores and crosstables, are available in the TCEC Archive. The archive also covers TCEC previous seasons.

TCEC S14

What about TCEC Season 14? It starts today: TCEC Season 14 – Let’s get ready to !boom (chessdom.com). The announcement began,

Season 14 of the Top Chess Engine Championship, the premier chess software competition, starts this Monday, November 12th, at 16:00 CET. It will involve the strongest chess programs and neural networks in the traditional division system. A total of 36 engines with ELO 3000+ will divided into five divisions and a Superfinal.

The action can be followed on TCEC - Live Computer Chess Broadcast.

CCC / CCCC / CCCCC

Some of those 'C's stand for Chess.com. The others stand for Computer Chess Championship, as in CCC: Computer Chess Championship, which happens to be the main page for the event.

Chess.com hasn't issued any news updates since the links mentioned in my 'Catching Up' post. Stage three, the final stage of the CCCC2 blitz competition, started today. The results of stages one and two must be somewhere, although I couldn't find them. I could work out stage one for myself from the PGN file (NB: Similar format for 'ccc1/stage1', etc.):-

https://newman.chess.com/archive/ccc2/stage1/games.pgn

Four engines have reached the final stage -- Houdini, Komodo, Lc0, and Stockfish (which jumped quickly into the lead) -- the same four that reached the semifinal stage of the TCEC Cup. For an overview of the engines that competed in stage 1, here is a colorful graphic from Chess.com's most recent news item. The little green and yellow trophies show the number of TCEC/CCC(C) competitions won by the top engines.

I calculate that the games are played at the rate of about three games per hour, which means 72 games per day. With 600 games total, the event should be over some time next week.

11 November 2018

'The World Is Watching'

Today is the rest day after the first two games of the 2018 Carlsen - Caruana title match. For the blog's November video let's look at what's available from YouTube's World Chess channel, the organizer of the event. So far the channel offers:-

  • Opening Press Conference
  • Opening Gala
  • Game 1
  • Game 2

Here's the start of the first game, although you would never know it from the title or the description of the video.


World Chess Championship London 2018 (34:01) • 'Streamed live on Nov 9, 2018'

The description of the clip says only,

First 30 minutes of the World Chess Championship

There is no game number, no link to the official site, no explanation of why only the first 30 minutes are shown or how to watch the rest of the game. C'mon World Chess! You are supposed to be the experts in marketing chess to the entire world. You can do better than this. The banner in front of the players promises, 'The World Is Watching'. Make it happen!

If you've ever wondered about GM Caruana's nationality, he explained it during the 'Opening Press Conference'. In response to the question,

You have changed nationality a couple of times. Do you feel American or Italian or what do you feel?

he said,

I never changed nationalities. I had dual nationality from birth. I changed federations when I was 13. I was a kid, I was living in Europe, and this was mainly to get back to my roots -- my mother is Italian. When I changed back to the U.S. I was in my 20s. This was a personal decision and I feel connected to both countries. I would like to represent both, but only one is possible. I live in the United States and represent the U.S. now. I feel very much an American, but I cherish my Italian roots.

In case you're wondering, the official site is london2018.worldchess.com. The first two games both ended in draws. Good luck to both players!

09 November 2018

GPU Pricing

Let's say I want to buy one of those machines featured in my previous post, GPU Configurations. How much money will I have to shake out of the piggy bank?

The 'Lambda Blade' link I referenced leads to another page Blade : Customize. There are three configurations, shown below, where the model in the previous post is the 'Basic' version.

The pricing for the three models is:-

Basic • $20,788.00
Premium • $25,852.00
Max • $51,958.00
also: Customize

In other words, I can buy a new car or I can buy a 'Deep Learning' rig. I'm sure my wife would prefer a new car.

The 'Customize' option starts with the 'Premium' model and allows to add/subtract options to determine a final price. The numbers are consistent. If I decrease 'GPUs' and 'Memory' to arrive at the 'Basic' model, the price is the same for that model.

What exactly are the technical tradeoffs for the different prices? I'll look at that in another post.

08 November 2018

Match Tales from the Twittersphere

The much anticipated 2018 Carlsen - Caruana World Championship match starts tomorrow and journalistic protocol requires at least a perfunctory nod. What is Twitter saying about it?

On top of writing about the match itself, the Twitter angle gives me the chance to try another of the Algorithmia (AI?) services that I first used in From Black & White to Brown & Blue (October 2018). The particular service is Retrieve Tweets With Keyword (algorithmia.com). I fed it the three keywords 'chess carlsen caruana' and it spit out text examples of the 500 most recent tweets/retweets on the subject, in reverse chronological order, the earliest from yesterday. The 'retweets' aspect means that the same tweets are repeated over-and-over-and-over, most of which are a link to a source other than Twitter.

Of all the standalone tweets that I looked at, the most interesting is shown below. I initially tried to use the tweet embedded directly into this post, but the dimensions of the photo broke the blog container. I couldn't see a way to fix this quickly, so I resorted to a screen capture and shrank it.


https://twitter.com/seaningle/status/1060533865259954177

As for linked articles, the Twittersphere gave me three from Chess.com:-

Of the non-chess sources flagged by Twitter, another handful appeared over-and-over-and-over:-

As for the oldest of the 500 tweets/retweets, the distinction goes to another article from Chess.com:-

As interesting as the Algorithmia exercise was, I could have saved a lot of time with a straight Twitter query:-

twitter.com/search?q=carlsen+caruana.

Next time I'll just do that.

06 November 2018

Chess in Antarctica

That's the South Pole in the crosshairs on the map of Antarctica. The word 'McMurdo' is to its left.


May 1964, Source unknown

The accompanying newspaper clipping said,

CHESS ON ICE • Static Stops Play

A chess tournament between Christchurch [New Zealand] players and Americans at McMurdo Sound, Antarctica, had to be adjourned with 10 games only half-way through when radio conditions deteriorated yesterday afternoon.

The play was tapped out on teleprinters in the communications room at the American base at Harewood and on teleprinters at McMurdo Station. But as the sun sank lower the radio frequency had to be changed, and difficulties at the Antarctic end began to slow up the games. The Americans suggested an adjournment to Sunday, June 14.

"It's a bit hard to tell," said Mr. A. S. Hollander, president of the Canterbury Chess Club, "but it looks as if we are leading." Chess is one of the very few ways in which the winter population of McMurdo Sound can participate in the life of the outside world, and Mr. Hollander thinks the Americans would like to play a lot more chess in future. "I'm sure there'll be a return match at least," he said.

My guess is that the photo was taken on the New Zealand side of the radio match. Why are there two boards in front of each player?

05 November 2018

So Many Ideas, So Little Time

In last week's post, Happy 10th Anniversary, Stockfish!, I noted,

The name of choice for the commemorative release was 'Stockfish X', but with only a few days to go, it doesn't look like anything special is going to happen.

And nothing did happen. Let's continue with the Stockfish forum, FishCooking -- first seen on this blog three years ago in Chess Engines : FishCooking (October 2015) -- and look at some of the recent posts that attracted discussion from the Stockfish community *and* that provide some insight into the group's working methods. First up are a couple of forum posts about selecting a version for the TCEC final match, which is currently underway and nearing its conclusion.

  • 2018-10-20: Executable for TCEC superfinal? • 'Apparently we haven't submitted an update to play in the TCEC Superfinal yet. I suspect it is too late now, but perhaps the senior devs could prepare one or comment on whether the e.g. abrok one should be used and contact TCEC?'

  • 2018-10-21: Request for binary for TCEC superfinal • Marco Costalba: 'We have few hours left before TCEC deadline is met to provide a binary for the superfinal. We would need a fast binary of the current master within this evening (European time) and I will send to TCEC people. If not, we will fallback on current TCEC binary. If someone has a better idea, please share it.'

The open source nature of Stockfish development presents many challenges.

  • 2018-10-17: Discussion for optimizing methodology • 'SF has been able to progress with the same 7(?) year old recipe and without doubt still is, but in my opinion with some adjustments in accordance to its current situation and needs it can be helped tremendously'

A technical issue related to evaluation has been under investigation for months.

  • 2018-10-03: First results of contempt tests • 'Actually on STC it seems that optimal contempt value is somewhere between 30 and 70. I will try to build the same diagram for LTC and if it shows the same stuff then we can probably conclude that for TCEC divP optimal value of contempt will be much higher than default 21 - closer to 50 because field there is much weaker than SF9, in fact, every single engine there is weaker than SF9.'

As those excerpts show, the forum members use so many acronyms and so much jargon that they might easily dissuade casual chess engine fans from trying to understand what they are discussing. Here are a few terms that are worth knowing. These first two acronyms I covered in a recent post titled Catching Up with Engine Competitions (October 2018):-

CCCC = Chess.com Computer Chess Championship
TCEC = Top [formerly 'Thoresen'] Chess Engines Competition

The next two acronyms are used throughout the forum. If you don't know what they mean, there will be many discussions that you won't understand. The terms are key elements of a testing strategy that starts with STC games and, if no problems are discovered, continues with LTC games.

STC = Short time control
LTC = Long time control

This next term is not specific to chess.

SPRT = Sequential Probability Ratio Test

It is a type of statistical analysis that is explained in Wikipedia's Sequential probability ratio test. The next acronym stems from a web domain that is used for administration, e.g. Stockfish Development Versions.

ABROK = abrok.eu

The word 'contempt' in the last thread listed above is a current topic of experimentation. In Contempt Factor (chessprogramming.org), the wiki defines it as:-

The Contempt Factor reflects the estimated superiority/inferiority of the program over its opponent. The Contempt factor is assigned as draw score to avoid (early) draws against apparently weaker opponents, or to prefer draws versus stronger opponents otherwise.

In other words, even if the evaluation shows a small advantage for the opponent, treat it as equality (a 0.00 evaluation). The typical debate is about what numerical value should be used. If it's too large, you risk losing drawn games; if too small, you draw games that you might win.

Another thread that has been running for nearly too years points to a useful tool for understanding Stockfish evaluations.

  • 2017-01-22: "Stockfish Evaluation Guide" tool • 'I developed tool where you can investigate each part of Stockfish static evaluation function. It is standalone single HTML page with javascript. Every evaluation term is rewritten in single small javascript function. You can setup any position with FEN or by moving pieces on chessboard and see how evaluation is computed and what is result and if possible to attach score to individal squares it is visualized on chessboard.'

The thread eventually points to Main evaluation (hxim.github.io). If I had the time, I would definitely take a closer look at it. I could say the same for many of the threads in the FishCooking forum.

04 November 2018

Elegant 19th Century Austrians

For this November edition of Top eBay Chess Items by Price (March 2010), my first choice would have been a painting titled 'Charles Schreiber, 1845-1903 French, "A Game of Chess" Oil'. Then I realized that it was exactly the same item that I featured three months ago in Chess with Two Cardinals (August 2018) -- same title, same description. This time the painting sold for US $1500 after one bid.

Instead of the cardinals for a second time, I picked the painting shown below. Titled 'A Game of Chess Antique Genre Oil Painting Josef Morgan (Austrian, 1839–1898)', subtitled 'Attributed to Morgan - Signed with Initials - Inscribed', it sold for GBP 620.00 ('approximately US $803.71', according to eBay) after receiving 37 bids from 14 bidders.

The description added little to the information given in the title:-

A very fine 19th century genre painting depicting an elegantly dressed couple playing chess in an interior which is presumably the work of Austrian artist Josef Morgan (Austrian, 1839–1898).

The frame (not shown here) carried a small plaque, 'J. Morgan 1894'. Another variant of the name is Jozef Belohlawek Morgan. Whatever the artist's real identity might be, we have another reminder that Chess Paintings Require Dogs (April 2011).

02 November 2018

GPU Configurations

In my previous post, GPU Specifications, I closed with:-

The emphasis in the video is on gaming, but the connection with AI is clear enough. I'll look at that in my next post.

This led me to a page, Deep Learning Server - 8x GPUs (lambdalabs.com), about a specific product, called the 'Lambda Blade'. The essence of the product is captured in the following image, the physical machine on the left, its component list on the right.

The essence of the 'What's Inside' column is repeated below. The small print in the image lists options for upgrading this basic configuration.

OS: Ubuntu 18.04 (Bionic)
GPUs: 8 x NVIDIA GeForce GTX 1080 Ti
CPUs: 2 x Intel Xeon E5-2650 (12 cores, 30M Cache, 2.20 GHz)
RAM: 128 GB DDR4 ECC
SSD: 2 TB (SATA)
HDD: 4 TB (SATA)

This basic introduction leads to a 'Customize' page, which I'll look at in my next post.

01 November 2018

November 1968 'On the Cover'

Chess magazine covers aren't reserved exclusively for World Champions and top grandmasters. Sometimes they feature prominent members of the B-team. Here we have two of them for the same month.


Left: 'I.S. Turover, "The Great Chess Lover"'
Right: 'Artist in Chess'

Chess Life

Last month's October 1968 'On the Cover', featured a report by TD George Koltanowski on the 1968 U.S. Open. This month's cover led to a continuation of that report.

Attending the U.S. Open at Snowmass from the start to the bitter end was none other than the "Grand old Man" (if I said "young" I would be more correct), I.S. Turover of Washington, D.C. The septuagenarian was in the thick of things, no matter at what time the games were adjourned in the evening or continued in the morning. And when the masters assembled at the special dinner given by I.S., we heard wonderful tales of the chess past from E. Schuyler Jackson of New York and from I.S. himself. Here's one of them:

The tale, omitted here, was a story about Turover meeting Alekhine and his wife.

I.S. Turover was born in the same city as M. Najdorf, in [Warsaw] Poland, and came to New York in 1912, where he played in his first tournament in 1913. At 17, he won the Championship of Washington, D.C. He won it altogether three times, gaining permanent possession of the Washington Post Trophy, retiring undefeated in 1928!

In business, he started as a jack-of-all-trades, and then started making good in the lumber business. Although he was blessed with success, he never forgot his great love: chess. Though well known in all charitable and civic circles as a great donor and supporter, he also supports chess. Only recently he participated in the sponsorship of Bobby Fischer's trip to the Natanya [Netanya] tournament in Israel, and he accompanied Bobby to the event.

Chess Review

Marcel Duchamp died at 81 last October 1st, in Paris. In the art world, he became the "stormy petrel," much as Nimzovich had been in chess, taking a leading part in rebellion against established conventions and beliefs and in promoting the skeptical school of Dadaism. As of today, however, he holds a revered place as a prophet of and leader in most or the recent experiments in art forms. In consequence, his obituaries have been impressive and featured as front-page news in the papers.

Most of the obituaries carry Duchamp's half-serious statement that he gave up painting for chess. In chess, he was a quiet and genial figure, playing a strong game and active mainly in New York state chess circles and in the London Terrace Chess Club in New York city. Duchamp was a Frenchman, became an American citizen; but, in art and in chess, he was truly international. • Photo by courtesy of the New York Times

This Chess Review would be the first issue of the last 12 months of the publication's existence.

30 October 2018

October Yahoos

The two Yahoo stories I spotted this month were both about the FIDE Presidential election, which I covered on my World Chess Championship blog:-

Both Yahoo stories used similar photos of outgoing FIDE President Kirsan Ilyumzhinov to illustrate the headline. The logo for the Russia news agency, Interfax (wikipedia.org), is seen in the background.

2018-10-03: Global chess body elects new head amid claims of Russian meddling (yahoo.com; AFP). At the time I captured the Yahoo introduction shown above, the headline and the partial paragraph beneath the headline didn't match. The headline said that Dvorkovich had been elected. The lead said the 'federation is set to vote'. Here is the entire lead sentence and first paragraph:-

The global chess federation is set to vote for a new president on Wednesday, replacing its veteran, highly controversial Russian chief, amid allegations of undue influence exerted by Moscow. The race to head one of the world's largest sporting federations will see former Russian deputy prime minister Arkady Dvorkovich run against current World Chess Federation (FIDE) deputy president, Georgios Makropoulos, who is Greek, and British grandmaster Nigel Short. Whatever the result, the election will bring to an end the more than two decades-long reign of Kirsan Ilyumzhinov as president of the body uniting the world's 188 national chess federations.

Now the lead paragraph on the same link says,

Batumi (Georgia) (AFP) - The global chess federation elected former Russian deputy prime minister Arkady Dvorkovich as its new president Wednesday, replacing a colourful but controversial chief amid allegations that Moscow was putting pressure on national federations.

I don't like online news stories that change from one moment to another. There's something sneaky, bordering on deceptive, about them. The second story was more straightforward.

2018-10-03: World chess gets first new president in two decades as Putin loyalist takes power (yahoo.com; The Telegraph).

The World Chess Federation elected a new leader today bringing to an end the colourful two-decade rule of a Russian politician who claims he was abducted by aliens. Arkady Dvorkovich, Moscow's preferred candidate, was installed as the new king of Fide after delegates from 182 nations voted at its congress in the Georgian resort city of Batumi. Mr Dvorkovich, the chief organiser of the Fifa World Cup, replaces the eccentric businessman-turned-politician Kirsan Ilyumzhinov who reigned supreme over the chess world for 23 years.

There was one other significant story that appeared on Yahoo, but which I didn't see in their headlines. It could be even more important to the development of chess than the election stories above.

2018-10-20: Meet the first American since 1972 to compete for the world chess title by Daniel Roberts. The same writer was behind two previous posts on this blog:-

Let's hope that 'November Yahoos' will be filled with stories about next month's 2018 Carlsen - Caruana match, although the often repeated 'first American since 1972 to compete for the world chess title' is a disservice to Gata Kamsky. For more on the subject, see Gata Kamsky taking objections to the claim that Caruana is the 1st American since Fischer to play for World title (reddit.com).

29 October 2018

Happy 10th Anniversary, Stockfish!

In last week's post, Talking About Chess Engines, I noted,

The two stars of the chess engine world are Stockfish and Leela. [...] Both engines are open source and both have forums where their many fans can discuss current trends. I really should spend some time on both forums to catch up with their technical developments and just might do that.

The last time I spent time on Stockfish's FishCooking forum was Stockfish in a Straitjacket? (March 2018), so I had tons of material to review. It was time well spent. One thread informed,

  • 2018-07-23: Stockfish 10th Anniversary • 'Is there any chance for next Stockfish 10 to be released 2 November 2018 as of 10 years of Stockfish first release back in 2008 and for name to be "Stockfish 10th anniversary"?'

The name of choice for the commemorative release was 'Stockfish X', but with only a few days to go, it doesn't look like anything special is going to happen. Another thread pointed to a series of graphs showing Stockfish's recent rating progress. This particular graph shows progress since the end of 2015, nearly 200 rating points.


Fishtest/wiki: Regression Tests

As I write this, Stockfish is playing the TCEC Season 13 final against Komodo and is leading +5-1=36, with another week or so to reach the match target of 100 games. The TCEC gives the engines' current respective ratings as 3519 and 3475.

The conventional wisdom is that people shouldn't play important chess games on their birthdays. I doubt that it matters much to engines.

28 October 2018

Chess and the Behavioral Sciences

We're approaching the second anniversary of the series on The Sociology of Chess (November 2016), and I still haven't introduced what appears to be the most important work related to the subject, Players and Pawns (uchicago.edu), by Gary Alan Fine, subtitled 'How Chess Builds Community and Culture':-

A chess match seems as solitary an endeavor as there is in sports: two minds, on their own, in fierce opposition. In contrast, Gary Alan Fine argues that chess is a social duet: two players in silent dialogue who always take each other into account in their play. Surrounding that one-on-one contest is a community life that can be nearly as dramatic and intense as the across-the-board confrontation.

This video, from the American Folklore Society, might have nothing to do with chess, but it does serve as an introduction to the author and his work.


Gary Alan Fine, "The Folklore of Small Things: Tiny Publics and Realms of Local Knowledge" (1:48:09) • 'Published on Jun 19, 2017'

The description of the video says,

Gary Alan Fine (Northwestern University and the Center for the Advanced Study of the Behavioral Sciences), delivered the 2010 Francis Lee Utley Memorial Lecture of the AFS Fellows at the American Folklore Society Annual Meeting in Nashville, Tennessee on Friday, October 15.

Abstract: "To understand contemporary society folklore requires a robust theory of how small groups motivate the creation and retention of tradition. The establishment, ordering, and expansion of any culture depend on groups with shared pasts and futures, that are spatially situated, and that depend on common references. Folk cultures arise from interaction scenes, linked to a field of activity. Within complex societies, specialized groups fulfill a set of instrumental tasks within a complex division of labor. As a result many group cultures are linked to the presence of knowledge specialists: experts who serve as brokers for external, lay publics. These groups constitute epistemic communities linked to focused knowledge realms, achieving essential societal ends in the absence of general knowledge."

While writing this post I only watched the introduction to the lecture. Once it's posted, I'll watch the rest. If there is anything about chess, I'll mention it in a follow-up paragraph.