Computer science at the University at Buffalo New York- State University

When it comes to cheating, chess might seem all but invulnerable. After all, the board and its pieces are out in the open for all to see. But an eruption of recent scandals has made it clear that cheating - fuelled by powerful computer programmers that play better than people do, as well as sophisticated communication technologies - is becoming a big problem for world championship chess. Last year, the French Chess Federation accused three players of colluding at the Chess Olympiad in Russia in 2010 by using coded text messages and a signaling system. The federation banned the players for five years, though the ruling is under appeal. When they come under suspicion, how can officials determine whether they are cheating? That is where Kenneth W Regan comes in. An associate professor of computer science at the University at Buffalo New York- State University, Mr.Kenneth W Regan has been researching the problem for the last five years. To detect possible cheating, Mr.Kenneth W Regan wanted to create a model of how often the moves of players of varying ability match those of chess programmers, so he began building a database by downloading and analysing games dating to the early 19 {+t} {+h} century.

In each game, Mr.Kenneth W Regan had the computer evaluate each position in single-line mode to a depth of 13 ply (six or seven moves by each player). To date, Mr.Kenneth W Regan has analysed nearly 200,000 games, including all of those from the top 50 major tournaments in history. Mr.Kenneth W Regan also has analysed 7,000 games to create models of different player abilities. To test someone for cheating, Mr.Kenneth W Regan runs that player's relative skill ranking, known as an Elo rating, against the comparative model. The research has yielded some interesting things about how chess programmes work, particularly Rybka, the strongest of the commercially available products. For example, in situations in which the evaluation of the position is particularly unfavourable, the programme can get stuck looking for a solution. "I think that 47 hours is my longest stall," Regan said. He has also discovered that the way people play has evolved. According to his analysis, the player now ranked No. 40 in the world plays as well as Anatoly Karpov did in the 1970s, when he was world champion and was described as a machine. Regan says his models are at a stage where they can be used only as support in cases in which cheating is alleged.

Source: The Economic Times