Computer programme built to check cheating in chess

Chess blog for latest chess news and chess trivia (c) Alexandra Kosteniuk, 2012

Hi everyone,

We just found this very interesting article on chess and cheating in The New York Times. An associate professor of computer science at the University at Buffalo who is also an international master at chess, Dr. Kenneth W. Regan has been researching the problem for five years.

He was fairly certain that anyone using a program to cheat would have it set in single-line mode — in which the program quickly selects a possible move , then runs through a sequence of moves to evaluate its soundness. That is efficient, but not thorough.

Dr. Regan decided that he also needed to have his programs running in multiline mode so that he could see where and why the programs changed their evaluations. That takes much longer.

He wanted to create a model of how often the moves of players of varying ability match those of chess programs, so he began building a database by downloading and analyzing games dating to the early 19th century.

At the Canadian Open last year, a player whose rating was 2,100 (a candidate master) beat three international masters, whose ratings are usually at least 300 points higher.

After analysing the games, Dr. Regan said, “I was able to prove that his intrinsic rating was in the 2,100s and the international masters had just played poorly.”

You can read the full article at this link and a related article here.

From Alexandra Kosteniuk's
www.chessblog.com
Also see her personal blog at
www.chessqueen.com