In my post about the Jeopardy-winning supercomputer Watson, I remarked that Chess Grandmaster Garry Kasparov “wouldn’t last five minutes against Ken Jennings.” That was no slight against Kasparov, who may well be the finest player in the entire history of Chess, and is proving to be a charismatic political figure in post-Communist Russia. It’s just that the skill set and knowledge base needed to succeed at Jeopardy is far more diverse and complex than that required to succeed at Chess.
Although Kasparov won his first match against IBM’s Chess-playing computer Deep Blue, he lost the rematch, and didn’t handle the loss very well. Obviously, he’s thought quite a lot about the man-versus-machine contest in the ensuing years, and he offers some interesting comments about Watson in The Atlantic:
My concern about its utility, and I read they would like it to answer medical questions, is that Watson’s performance reminded me of chess computers. They play fantastically well in maybe 90% of positions, but there is a selection of positions they do not understand at all. Worse, by definition they do not understand what they do not understand and so cannot avoid them. A strong human Jeopardy! player, or a human doctor, may get the answer wrong, but he is unlikely to make a huge blunder or category error—at least not without being aware of his own doubts. We are also good at judging our own level of certainty. A computer can simulate this by an artificial confidence measurement, but I would not like to be the patient who discovers the medical equivalent of answering “Toronto” in the “US Cities” category, as Watson did.
Read the whole thing.
h/t: My wife (again).