Cover Story / Ken Regan
ants played on boards larger than 8x8 are), but it shares two characteristics: 1) it is practically impossible to prove a solution—for example, to prove a win or draw for White from the initial position; and 2) we can quickly verify a solution— whether or not a particular chess position is a checkmate. The main difference be tween chess and NP-types is that the solution to chess is theoretically possible, whereas solutions to NP-type problems currently are not. In one way, however, chess can be marked more difficult than NP-types, because with NP-types one can theoretically verify a solution at the start if there is one. To find the solution to chess, one can only compute deeper and deeper. Claude Shannon, the father of informa tion theory, in his famous paper “Pro - gram ming a Computer for Playing Chess,” estimated the number of possible unique
chess positions to be roughly 1043. “It’s impossible to unpack the complete game tree,” says Regan. “It’s so large that if those bits were placed in an efficient memory device the size of a room, that room would collapse into a black hole.” Regan classifies chess as a Deep problem, “One where I can describe the complete set of rules in a small amount of information, but where unpack ing the information will take a long time.” Chess engines continue to improve at about 20 Elo points per year. If Regan’s
estimate of perfect play at 3600 Elo is true, then they will arrive there within a few decades. Regan believes they already play perfectly on occasion, if given enough time to “think.” A chess computer with a good enough algorithm and fast enough processor does not need to store 1043 positions to play with the same skill as a computer that does. To understand how such an astonishing feat is possible, consider how it’s possible for a human to play perfect tic-tac-toe without having to store the complete solution to tic-tac-toe. There are 256,168 possible different games of tic- tac-toe, but a little smarts reduces this number to 230 strategically important positions.
IN SEPTEMBER, 2013, HARVARD University hosted the one-day New England Symposi - um for Statistics in Sports, and Regan decided to attend on a stopover while on his way home from another conference. “I’m not going for the talks so much as to hobnob and button hole people,” he wrote to me a few weeks before the event. We met at the bar of the Grafton Street Pub, a crowded restaurant near Harvard Square, for the symposium’s social hour. The din of Saturday night beset normal conver- sation, and I found Regan leaning into the voice of Eric Van, a 50-something statis- tician who consulted for the Boston Red Sox between 2005 and 2009. Van was explaining to Regan how the Sox needed to shuffle their lineup to win the World Series, a task Van helped the team achieve in 2007 and one in which they eventually accomplished again a month after this get- together. Regan had come to the conference to form connec tions, and Van’s attachment to baseball made this one particularly sweet. But the whole bar scene injects a bit of
anxiety into Regan’s body language. He blinks hard at times and chews his gum vigorously. (Later Regan would tell me, “Chess got me comfortable in an adult world. I was able to step right off the boat my first year as a graduate student at Oxford and feel confident.” It was during this time at Oxford when Regan also met his wife.) Regan offered to buy us drinks, in a tone of voice that implied this wasn’t a question he often asked but which he felt was the obligatory thing to do. Nobody
accepted. So he bought a pizza to share, and we moved to a quieter spot. Van battles attention deficit disorder
and narcolepsy, and now spends his time as a private scholar who researches these ailments and works on a theory of con - scious ness. The conversation turned to the intersection of cognitive science and chess, which naturally led to a discussion about the hypothetical Chinese Room Thought Experiment first proposed by philosopher John Searle. In this experiment, an English-speaking
person sits in a locked room. After a question written in Chinese is slipped under the door, the person follows rules on a flowchart that describes how to write an answer in Chinese. The person then slips the answer back under the door. It would appear to people outside the room that there is an intelligent, native Chinese
1993 WORLD OPEN (PHILADELPHIA) An unrated player using the pseudonym “John Von Neumann" scored 41
1999 BOBLINGER OPEN (GERMANY) 55-year-old and 1925-rated Clemens Allwerman scored 71
⁄2
/9 in the Open
section, including a draw against GM Helgi Olafsson. "Von Neumann" was disqualified after he refused a request by a suspicious tournament director to solve a simple chess puzzle.
⁄2 /9 to win the tournament ahead
of multiple titled players. Subsequent analysis using the then-current Fritz engine roused tremendous suspicion, but Allwerman was never disciplined.
2006 SUBROTO MUKERJEE MEMORIAL OPEN (INDIA)
1933-rated Umaket Sharma caught with a Bluetooth device in his hat. Sharma was suspended 10 years by the All India Chess Federation.
2006 WORLD OPEN (PHILADELPHIA) 1974-rated Steven Rosenberg disqualified for using a Phonito, a hearing aid-sized device that slips into the ear, and which can send and receive wireless communications.
2010 FIDE OLYMPIAD (KHANTY-MANSIYSK)
19-year-old GM Sébastien Feller, GM Arnaud Hauchard and IM Cyril Marzolo caught performing an elaborate move-relaying scheme. Marzalo, who would be home at his computer, would text Hauchard in the playing hall, who would then communicate moves to Feller, based on where he was standing. According to Wikipedia, the French Chess Federation’s suspension of these three players was revoked, but Feller is currently serving a 33-month suspension from FIDE.
2011 GERMAN CHAMPIONSHIP
FM Christoph Natsidis used an engine running on his smartphone, while in the bathroom. Natsidis was disqualified from the tournament.
2012 ZADAR OPEN (CROATIA), 2013 BLADOEVGRAD OPEN (BULGARIA), NAVALMORAL OPEN (SPAIN)
26-year-old Borislav Ivanov scored 6/9 at the Zadar Open, including wins over four grandmasters. When Ivanov refused inspection at Bladoevgrad and Navalmoral after suspicious behavior and wins over multiple grandmasters, he was forfeited both times. Eventually Ivanov was suspended for four months by the Bulgarian Chess Federation.
www.uschess.org
27
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76