I just read the paper the New Scientist report is based on, and it's quite interesting, despite a few annoying inaccuracies in it (Maven did not beat Ben Logan in 1998, and to say that it beat Adam decisively after so few games played is misleading. Also, Adam was not yet World Champion; later, IMINO is played at 8D, not 8E).
I don't think that "playing dirty" is an appropriate descriptor for making inferences. But in a way, simming programs do play dirty in that they give information to the computer that is impossible for humans to attain. Is it true that Jim Homan's program raked in 3 million dollars? I find that almost unbelievable. The paper is at: http://reason.cs.uiuc.edu/eyal/papers/scrabble-ijcai07.pdf Joel W. --- In [email protected], Darryl Francis <[EMAIL PROTECTED]> wrote: > > There's a short piece in the new New Scientist journal (13 January) > entitled "Robot learns to play dirty". See: > http://tinyurl.com/yxw4uh > > Part of the article says: > "Adding in this 'opponent modelling' greatly improved the program's > game, allowing it to beat Quackle, one of the best conventional > Scrabble programs, by five points on average." > > Would the Quackle authors care to comment? > > > > > > > > > ___________________________________________________________ > New Yahoo! Mail is the ultimate force in competitive emailing. Find out more > at the Yahoo! Mail Championships. Plus: play games and win prizes. > http://uk.rd.yahoo.com/evt=44106/*http://mail.yahoo.net/uk >
