Began reading Game Changer: AlphaZero’s Groundbreaking Strategy and the Promise of AI,
by GM Matthew Sadler and WIM Natasha Regan
recently. I have only read a couple of chapters and have no intention of writing a review because the book has been reviewed by almost everyone but this writer.
For those of you who wish to read a review before purchasing the book I heartily recommend the one by GM Jacob Aagaard
in the best Chess magazine in the world,New In Chess, issue 2019/3. Kudos to the people at NIC who make the decision as to what goes into the magazine, and what stays out. Aagard was nice about ripping the authors new ones, writing, “Game Changer is an interesting but often also frustrating read.” In addition he writes, “However, the structural problems the book suffers from are certainly to do with the two authors, two voices and at least two different directions.” There is more but I will not dwell on it other than to say reading the review caused me to purchase the book after reading, “The book gets into a better flow over the next hundred pages, before becoming coherent over the last 250+ pages that look seriously at AlphaZero’s games,” which is the basic reason for buying the tome.
I attended the Chess club Thursday night at the local Barnes & Noble bookstore. Because of my age and having only recently sufficiently recovered from illness I informed the TD I would be willing to act as a “filler” in the event there were an odd number of players and would only play in the first two rounds.
Having attended the previous week, the first time I had made it in some time, a few new players were noticed, which the gentleman who runs the club attributed to the recently finished match for the Human World Chess Championship. Most, if not all, of the players who attend are so hungry for a game they play “skittles” games before the G/15 event begins. There was a “newby” who caught my eye because he was wearing sandals during winter. He looked as though he would have fit in at Woodstock in 1969, so I spoke to the young man, saying, “You gotta like a guy who refuses to give in to winter.” His name was Dawson and he was ready to play, someone…anyone, so we sat down for a game after introductions. I had the white pieces and opened with 1 e4. He responded with the French move of 1…e6. After playing the standard 2 d4 he answered with 2…d5, whereupon I advanced my pawn to e5 on my third move. My opponent stopped to cogitate a few moments before playing 3…Nc6 with obvious trepidation, which showed when he kept his finger on the Knight after placing it gingerly on the square. As he did so I took a good look at him while thinking he appeared about the same age as I was when first visiting an official Chess club. He finally removed his finger from the Knight. I continued looking at the young man, wondering if I should say anything…Before speaking a particular scene from one of my favorite movies flashed in my mind:
When he looked up from the board I said, “At the Chess Club we do not, ever, hold our finger on a piece. When you decide upon your move, make it like you mean it and place it firmly on the square with deliberation, and immediately remove your fingers from the piece.”
The young fellow was somewhat taken aback, but gathered himself quickly and nodded in assent. I continued, “Are you playing in the tournament?” He said he was not. “Then I suggest you spend some time watching these gentlemen play, paying particular attention to how they move their pieces.” Again, he nodded. I did not have to mention it again.
Granted, I am no longer the player I was earlier in my life, and having played over many of the games from the recent World Senior Chess Championship,
(http://www.wscc2018.european-chessacademy.com/index.php/en/) I realize how much of a decline there is for an old(er) player, especially in the 65+ section, which is now my category. That said, the young fellow played a decent game, developing his pieces in the opening without any extraneous pawn moves or outright blunders. We arrived at about an even position in the early middle game, before he made a mistake, moving his a-pawn aggressively, but weakening his b-pawn in the process. I secured my b-pawn by playing a3, then picked off his undefended b-pawn. A few moves later there was a tactical skirmish in which I came out a piece ahead, and he sort of went downhill from there. The game ended in mate by my newly minted Queen protected by a lone Knight.
“You played very well, young man,” I said. There were a couple of players watching the game and they seconded my remark. He said graciously, “I appreciate your saying that, sir.” We talked and I learned he was twenty years old, the same age as was I when I first went to the Atlanta Chess club. He mentioned coming because he was beating the players with whom he had been playing and wanted better competition. Wondering how he could play such a decent game I asked if he read any Chess books. “Not really,” he said. “But I’ve been on Chess.com and watched many YouTube videos.”
The tournament began and I was not needed, fortunately. This gave me an opportunity to watch some of the action, talk with some of those who come and play without playing in the tourney, and those who come to simply “hang-out.” It was immensely enjoyable. I watched Dawson play one of the young players who is not a member of the USCF (“It costs $30!”) but comes to play skittles. Dawson was a piece down but came back to win the game.
After becoming a Senior I began staying home at night for a reason. Although exhausted after being at the Chess Club I was unable to sleep soundly and the next day, Friday, was not one of my better days, so I took it easy and relaxed, spending much time reading, and listening to programs via the internet.
Fortunately, Saturday was a totally different story. I read while having my first cuppa joe. After breakfast the web was surfed. Chess is usually saved for last and one of the sites I visit every day is GM Kevin Spraggett’s
website (http://www.spraggettonchess.com/). He has a “Chess News” scroll, “What is Happening Today?” I clicked on the ones new to me and began reading. I read every article and there were many on AlphaZero. I even read an editorial by Garry Kasparov
I had intended on watching several videos by GM Matthew Sadler concerning the recent World Human Chess Championship games, but discovered videos at Chess24 in the article, AlphaZero really is that good (https://chess24.com/en/read/news/alphazero-really-is-that-good). I watched every video contained in the article superbly elucidated by GM Sadler. I was had by hook, line and sinker, after watching the first one, All-in Defence, “A true Najdorf brawl.”
The Najdorf was my first love. Like many others I played it because Bobby Fischer played the opening. With Bobby the Najdorf was an offensive defense.
While watching the Najdorf “brawl” I noticed another Sadler video over on the right and it looked like the position could have emanated from the Leningrad Dutch, my “second love.” I clicked on and, sure enough, it was a Leningrad! I was compelled to watch.
As if that were not enough I noticed a video by GM Ben Finegold, who married a woman in my home city of Atlanta and they opened the new Atlanta Chess Club & Scholastic Center. (https://atlchessclub.com/) The video is Capablanca Endgames with GM Ben Finegold.
I enjoyed Ben’s commentary while thinking, “I wish the internet existed in 1970.” How can young players, and even older players, not be far superior to those of my generation with tools like this, and the best players giving great advice away for practically nothing? Why would anyone pay someone to teach Chess?
In an email to Karen I wrote, “I did surf over to Twitch the other day to listen to the lonely Ben comment on the game. I was thinking it must be very difficult to do it alone for a long period of time…Ben the Maytag repairman…”
Karen replied, ” I don’t think he gets lonely streaming …. he seems to enjoy it and likes to talk a lot so it works out.” Ben talks a lot because he has something useful to say. He is like the old EF Hutton TV commercial. “When EF Hutton talks, people listen.”
Three new articles were found before writing this post at Spaggett On Chess and I intend on reading them later today, even the one by discredited economist and former GM Ken Rogoff:
I would like to bring your attention to an article by GM Jonathan Tisdall published January 23, 2018. It is one of the best, most insightful, Chess articles I have ever read. It begins:
Men and machines
Drama Kings
The middle rest day is the closest thing to a half-way mark at the 80th anniversary edition of the Wijk an Zee tournament. This year’s Tata Steel Masters (and Challengers) continues a brilliant tradition of offering more than the usual 9 rounds, and a cleverly composed mix of world stars and hungry, dangerous outsiders. For me, this event is the highlight of the tournament year, with only the Candidates offering comparable entertainment when it rolls around – though that is due to high stakes rather than careful and colorful organization. I like some extra rounds and some new faces, preferably crazed with aggression. This event always delivers.
Other headers tell you much about the content:
Top this?
Shakh shock
Older? Wiser
‘Winning’
Psychology
“This tweet sparked some interesting conversation. Yes, I suppose it is true that in a way Carlsen’s tireless technical determination is also a form of psychological warfare. But torture and fighting spirit are not such unique factors – there are of course occasional wizards at maneuvering or grinding, but these skills have also been part of the daily toolkit of gritty professionals, from those on the weekend circuit to the legendary Soviet school of endgame superiority. But great technical champions tend to spawn dazzling tactical successors; Capa to Alekhine, Karpov to Kasparov, and … Carlsen to ? Presumably someone who will play like AlphaZero, on a human scale, an UltraTal. That is the idea I was trying to summon up.”
Human frailty
Compensation
“And why is it that when world champions blunder a full piece, it isn’t quite a full piece?
Catching up on today’s #TataSteelChess On Magnus’s blunder: 17…f4 would be positionally catastrophic for Black if it didn’t win a piece.
One difference between humans and computers is that our strategic filters often trump our tactical filters at the worst possible moments.
— Jonathan Rowson (@Jonathan_Rowson) January 21, 2018
I am a huge fan of Rowson’s insights, and his unique examinations of chess matters psychological. His tweet sparked a few trains of thought – another component of his observation contains a kind of inherent law of compensation – even a blunder can result in practical chances if there is any price, particularly structural, to winning the material.”
Included this because I, too, am a huge fan of GM Rowson.
The human factor
Double trouble?
Game-changer
Muppets. Muppets!
The Candidates
So dangerous
The remarks contained under Psychology brought to mind a position from the 2018 Tata Steel Challengers tournament, aka, the ‘B’ group.
Provocative play by black, to say the least. It has been written that GM Viktor Korchnoi would sometimes play somewhat ‘dubious’ opening moves in an attempt to cause problems for his opponent, especially those who were known to be ‘booked-up’. The player of the white pieces has just made her thirteenth move, which completed her development, while black, Benjamin Bok, lags behind in development.
Houdini, a computer Chess program with the third highest rating, surprised the computer Chess world by making the TCEC Super Final against the favorite Komodo, then astounded the computer Chess world when it convincingly beat the Dragon. Houdini becomes the engine with most titles in the decade history of TCEC.
In an interview Robert Houdart, author of the champion engine Houdini, said, “It is a great feeling to win the Top Chess Engine Championship and be the engine with the highest number of titles. I’ve worked non-stop for the past two years to bring Houdini back to the top level, and I’m really happy that this has resulted in a new TCEC title, which is the equivalent of “world champion” status. Before the tournament I expected a close contest between Houdini, Komodo and Stockfish, and that’s exactly what we’ve got.” (http://www.chessdom.com/interview-with-robert-houdart-author-of-the-champion-engine-houdini/)
For those who can afford and simply must have the new new thing, the new and obviously improved version of the Escape Artist is already available for purchase at the official website. (http://www.cruxis.com/chess/houdini.htm)
Unfortunately for Houdini this could prove to be a Pyrrhic victory what with the announcement near the end of the tournament of AlphaZero’s devastating victory over Stockfish, winner of last year’s season nine TCEC tournament. AlphaGo’s ‘machine learning’ is already being incorporated into some Go programs for sale and it is only a matter of time until that same technique will be incorporated into future versions of Chess programs, which will destroy the current big three Chess engines unless they also do what Google’s Deep Mind has done. (https://deepmind.com/) For example, consider this exchange from the aforementioned interview:
Alpha Zero just defeated last year’s champion Stockfish 8. Your opinion on the paper published and the match that took place?
It’s fascinating and amazing, and at the same time very much expected!… We even discussed this during the interview with Nelson and the Komodo authors. It opens entirely new, astonishing possibilities for chess engines! I do hope Google will publish more details about their approach, so that the chess world in general and the computer chess world in particular can benefit from their achievement.
Season 11 is starting after the rapid and blitz. TCEC is becoming a league and Houdini starts in the Premier division. Do you expect more competition besides Komodo and Stockfish for the top spots?
The gap between the top 3 and the rest is quite big, I don’t expect any other engine to bridge it soon. Then again, 16 months ago everybody was talking about the “top 2” and the rest, nobody expected Houdini to make a comeback. Anything can happen, Season 11 should be fascinating!
Google’s AlphaZero destroys highest-rated chess engine in 100-game match
Thursday December 7, 2017
Chess changed forever today. And maybe the rest of the world did, too.
Chris Garlock writes the AGEJ and this is his take on the development:
“A little more than a year after AlphaGo sensationally won against the top Go player, the artificial-intelligence program AlphaZero has obliterated the highest-rated chess engine.
Stockfish, which for most top players is their go-to preparation tool, and which won the 2016 TCEC Championship and the 2017 Chess.com Computer Chess Championship, didn’t stand a chance. AlphaZero won the closed-door, 100-game match with 28 wins, 72 draws, and zero losses.
Oh, and it took AlphaZero only four hours to “learn” chess. Sorry humans, you had a good run.
That’s right — the programmers of AlphaZero, housed within the DeepMind division of Google (https://deepmind.com/), had it use a type of “machine learning,” specifically reinforcement learning. Put more plainly, AlphaZero was not “taught” the game in the traditional sense. That means no opening book, no endgame tables, and apparently no complicated algorithms dissecting minute differences between center pawns and side pawns.
This would be akin to a robot being given access to thousands of metal bits and parts, but no knowledge of a combustion engine, then it experiments numerous times with every combination possible until it builds a Ferrari. That’s all in less time that it takes to watch the “Lord of the Rings” trilogy. The program had four hours to play itself many, many times, thereby becoming its own teacher.”
“Think of the endless possibilities for silicon based cheating. It is going to be a challenge for the anti-cheating committees to compare games and catch someone who is using a new algorithm based on AlphaZero. The unique and paradoxical moves may not be comparable to known move selection by present programs. If the chips are small enough, virtually anything could become the device that can be used to generate moves quietly, stealthily, and effectively.”
The thing about writing a blog is that one never knows what an email will bring. After spending an inordinate amount of time in front of Toby, the ‘puter, yesterday learning how to insert diagrams, and then putting together the post in order to have something in which to insert them, I determined that today I would spend time with the Daniel Gormally book, Insanity, passion and addiction: a year inside the chess world, while playing over Chess games on an actual board with pieces one can feel, and possibly “working” on the openings intended for the Senior Championship of the Great State of South Carolina, which is only ten days away, by going to the CBDB and 365Chess. Wrong, Ke-mo sah-bee! An email from my friend Mulfish arrived at 11:42 am, upsetting the Bacon cart…
“Looking forward to the AWs take on AlphaZeros stunning win over Stockfish,” was the message. “What’s this?” I thought, wondering if Mike was referring to the TCEC Computer Chess Championship that is in the final stretch. “But Stockfish is not participating in the Super Final,” I thought. I therefore fired off an immediate response: “To what, exactly, are you referring?” His reply was, “Look in the all things Chess forum.”
Although there are not as many incoming as there were before taking a long break from blogging, I have received several emails directing my attention here and there, and they are greatly appreciated. Checking the AW stats today showed many people in countries other than the USA reading the AW. In particular I noticed that today, as every day, there is one, and only one, reader in the Maldives. Thank you, whoever you are, and feel free to send an email, as I am curious by nature.
Keep ’em coming: xpertchesslessons@yahoo.com
This is the post found on the USCF forum that prompted Mulfish to fire a salvo at the AW:
Postby billbrock on Wed Dec 06, 2017 9:16 am #321974
“AlphaZero learned to play chess by playing against itself. After just FOUR HOURS of self-learning, it was able to decisely (sic) defeat Stockfish 8.0! (EDIT: this statement is slightly misleading. See downthread.) (100 games match: +28 =72 -0)
What’s really impressive: Stockfish was calculating far more deeply than AlphaZero (at least in terms of nodes per second). AlphaZero is just “smarter.”
After reading only this I thought, “Whoa! This will change not only my day, but possibly the future course of history!” The more I read the more convinced was I of the latter.
Bill Brock provided a link to a PDF paper, Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm
(https://arxiv.org/pdf/1712.01815.pdf) which I read immediately, blowing my mind…
Every morning I read while drinking my first cuppa coffee, and today was no exception. Toby is not fired-up until time to sit down and eat breakfast. I check my email, then the quotes of the day, followed by the poem of the day, which was The Writer’s Almanac, by Garrison Keillor, but it has been discontinued, so I’ve moved on to Poem-a-Day (https://www.poets.org/poetsorg/poem-day) & The Poetry Foundation’s Poem of the Day (https://www.poetryfoundation.org/). Next I click on the Drudge Report in order to understand what the enemy is thinking, and doing. Then it is the newspapers in digital form, the NYT, WaPo, and AJC. For you readers outside the USA, that would be the New York Times, the Washinton Post, and the Atlanta Journal & Constitution. Then I check out the word of the day (https://www.merriam-webster.com/word-of-the-day), before heading to check what was on the nightly radio programs broadcast while I am sleeping, Ground Zero with Clyde Lewis (http://www.groundzeromedia.org/), and the Granddaddy of them all, Coast to Coast AM (https://www.coasttocoastam.com/). You may think that Chess comes next, but you would be mistaken. I check out The Hardball Times at Fangraphs (https://www.fangraphs.com/tht/). Then I check out what’s happening in the world of Go (http://www.usgo.org/).
Then it is time for Chess! My routine is to check in at Chess24 (https://chess24.com/en) first in order to learn if there is a new article I will want to return to after checking out Chessbase (https://en.chessbase.com/), where there is usually something interesting to peruse. (Today is no exception because the lead article is, How XiangQi can improve your chess, which will be read. https://en.chessbase.com/). During the TCEC Championships it is then on to Chessdom (http://www.chessdom.com/), where I click onto TCEC (http://tcec.chessdom.com/). And then it is on to the Chess Granddaddy of them all website, TWIC, aka The Week In Chess (http://theweekinchess.com/), which is Mark Crowther’s wonderful website which contains a Daily Chess Puzzle, which I attempt to solve, in hopes it will keep my mind sharp. Why was I writing all this?…Just kidding!
The point is that I read so long this morning (Why Bob Dylan Matters, by Richard F. Thomas; Cover Me: The stories behind the GREATEST COVER SONGS of all time, by Ray Padgett, who has a wonderful website (http://www.covermesongs.com/); and Murder on the Death Star: The assassination of Kennedy and its relevance to the Trump era, by Pelle Neroth) in order to finish the latter. The point being that by the time I got to the email by Mulfish I would ordinarily have already seen the momentous news.
The excellent article by Colin McGourty begins: “20 years after DeepBlue defeated Garry Kasparov in a match, chess players have awoken to a new revolution. The AlphaZero algorithm developed by Google and DeepMind took just four hours of playing against itself to synthesise the chess knowledge of one and a half millennium and reach a level where it not only surpassed humans but crushed the reigning World Computer Champion Stockfish 28 wins to 0 in a 100-game match. All the brilliant stratagems and refinements that human programmers used to build chess engines have been outdone, and like Go players we can only marvel at a wholly new approach to the game.”
Colin ends with: “And where do traditional chess programmers go from here? Will they have to give up the refinements of human-tuned evaluation functions and all the existing techniques, or will the neural networks still require processing power and equipment not easily available? Will they be able to follow in DeepMind’s footsteps, or are there proprietary techniques involved that can’t easily be mastered?
There’s a lot to ponder, but for now the chess world has been shaken!”
“Shaken?” More like ROCKED TO ITS FOUNDATION!
If games people play are to survive they will be something like that described in the novel I consider the best I have read, Das Glasperlenspiel, or Magister Ludi, aka, The Glass Bead Game, by Hermann Hesse. (http://www.glassbeadgame.com/)
Or maybe a book, The Player of Games, by Iain M. Banks, which is not only one of my favorite Sci-Fi books, but also one of my favorite book about games.
The stunning news also caused me to reflect on a Canadian Sci-Fi television program I watched, Continuum, in which mega-corporations dominate the world in the future as time-travelers fight one of the largest corporatocratic entities, SadTech, which sounds an awful lot like Google. (http://www.imdb.com/title/tt1954347/?ref_=nm_flmg_act_6)
The Brave New World is here. The Science Fiction books I read as a youngster are no longer fiction.
Armchair Warrior 