Checkers solved… Chess next??
A report has been released that a computer scientist from the University of Alberta has solved the game of checkers (8×8 version). The venerable game which has been a source of entertainment for families has now joined the ranks of tic-tac-toe. This is what Professor Jonathan Schaeffer claims on his website dedicated to his checker program Chinook:
Checkers is solved. From the starting position, black to play can only draw against a perfect opponent.
He also gave a present on 8 May 2007 at the University of Alberta on solving checkers. The podcast is available here. Interesting stuff! He talks about a legendary figure Dr. Marion Tinsley who was a long-time World Champion. Tinsley terrorized the opposition for decades until he was held by the Chinook program in a four-game “Man vs. Machine” match in 1992. The podcast seems to focus more on the marvels of Tinsley than Chinook. In the podcast, Tinsley even appeared to be the equivalent of Garry Kasparov in terms of his determination.
Schaeffer does say that the solution for chess is quite a ways off.
Checkers has roughly the square root of the number of positions in chess. Given the effort required to solve checkers, chess will remain unsolved for a long time, barring the invention of new technology.
That’s reassuring. 😕 Well… given the past several “Man vs. Machine” matches, it looks as if that time is short.
Read the report here!
As the article points out, chess is not next. But certainly the methodology used to allegedly solve checkers is the same one being used to develop increasingly stronger chess playing systems. It’s a squeeze strategy … massive databases of human developed opening theory, combined with ever larger, perfectly pre-calculated, endgame tablebases. Add an increased middlegame calculation horizon, enabled by faster computer processor speeds, and its easy to understand the enormous chess strength of today’s machines.
The most interesting point of the article for me, however, was the speculation on whether humans can reverse engineer the perfectly architected move sequences of the checkers computer. Thereby deriving better positional principles and technique guidelines that humans could apply in their games. I believe this is just what Garry Kasparov was able to do after his loss to Deep Blue II. Remember, he went on a winning tear that hadn’t been seen in top level chess since Fischer’s march to the world championship over 20 years earlier. Garry may have lost against the machine, but he gained a deeper insight into how to play chess.