Posts tagged Jeopardy
The Trivia Championships of North America were held recently at the MGM Grand in Vegas, and a big thanks to Paul Bailey for organizing it and about 100 other people for being such fun to hang out with.
On Saturday, I had the honor of hosting a roundtable discussion about IBM's human-crushing Watson trivia machine, with guests Ken Jennings (74-time Jeopardy! winner and Watson victim) and Ed Toutant (Millionaire winner and former IBM engineer). This is the three of us afterward, looking like frat brothers about to go on a bender.
Yes, we are all that orange.
On Sunday, I had the fun of playing the Alex Trebek role for the final rounds of the team championship. No crazy-lady-thief-chasing required.
I apparently left my chin in the green room. But no matter: congrats to Jonathan Hess, champion in multiple individual competitions, and to team champions Frank In Court.
Paul Bailey deserves huge praise in creating a counterpart to the European Quizzing Championships, and the event was a roaring success. See you next year in Vegas, and (hopefully) this fall in Bruges!
Thanks to my own 13 games of Jeopardy! and the book about it and all, lots of people (including the New York Times) have asked for my opinion about the whole IBM computer vs. Ken Jennings vs. Brad Rutter cage match.
I'll say here what I've said all along: when you're playing at million-dollar tournament level, all of the players know (or can rapidly figure out) nearly all of the correct responses, no matter how arcane. Trust me, I've been there.
IBM wouldn't unveil their computer unless they were sure it would function similarly.
Ultimately, the difference between winning or losing usually comes down just to the ability to time the milliseconds between the time Alex finishes the clue and one of the producers activates the signaling devices.
Since a computer can obviously react more rapidly and consistently than a human can, it will probably win unless it is programmed to have a human-like random delay of a few milliseconds before hitting the buzzer.
Judging from news reports about the warm-up game, the computer has no such delay, and can buzz in instantly.
So that's the computer's advantage: not knowledge, not processing speed, not math or game strategy… just reflexes. And so unless the buzzer-response is programmed differently in the games that count, I think the computer is likely to win.
Brad and Ken will both be still a heck of a lot more fun to hang out with after the game, though.
(Full disclosure: Brad and Ken are both friends of mine, so I have some emotional interest in seeing them do well against the cyborg. Also, I was invited to play the computer in its last warm-up round, but I couldn't make it since I was still getting over the typhoid I picked up in Tanzania. Y'know, as always, my life is so boring…)