April 28, 2009
IBM Computer To Challenge Humans On Jeopardy
Can man be outsmarted by machine in a game of Jeopardy? IBM says it intends to find out.
The computer firm on Monday announced that its scientists have been working for almost two years on a computing system known as "Watson" "“ named for IBM founder Thomas Watson.
Watson is being designed as a question-answering machine that aims to take on humans in a TV episode of the game show Jeopardy.
"The aim is to get Watson to think and interact in human terms," IBM's Dr David Ferrucci told BBC News.
"It will try to understand a user's question and intent and understand it at a rudimentary level and provide and accurate and confident answer."
Jeopardy will be a tough challenge for a computer since its clues often involve "analyzing subtle meaning, irony, riddles and other complexities."
"The most challenging aspect of this is that Watson has to know what it knows with utmost confidence," said Ferrucci.
"Otherwise if it buzzes in and gets the answer wrong that is bad on Jeopardy because you lose money and lose the game."
IBM has tread familiar territory in the past. In 1997, it developed a computer called Deep Blue that successfully beat world chess champion Garry Kasparov.
The company had to create a fast computer that could calculate 200 million chess moves per second.
For the Jeopardy challenge, Watson will be given the questions as electronic text, while human competitors will be reading and listening to the question given by the show's host Alex Trebek.
Watson will not be connected to the Internet during the show, instead it will have to draw upon the previously stored knowledge.
"It requires critical thinking. It requires a whole lot more than knowledge alone," said Jeopardy's executive producer Harry Friedman.
"When people hear about this project, they say isn't this Google on steroids? No, it's not."
The computer may have a formidable challenger in all-time Jeopardy champion Ken Jennings. The show's producers hope to get Jennings to take part in the episode.
"Ken is the best and no-one gets close to his achievement. This is just a great test for Watson," said Ferrucci.
He added that while the approach is novel, Watson is intended for more than fun and games.
"This is about giving enterprise faster access to more precise answers," said Ferrucci.
"While business might be satisfied with the classic search approach because it gives you plenty to read, the thing that haunts you when you are making a decision is what is out there that I am not reading."
"Watson will look at all the information, including the most remote that might impinge on any answer you might make building more and more confidence into that content to help support your decision."
On the Net: