February 17, 2011

IBM’s ‘Watson’ Defeats Human Contestants On Jeopardy!

Displaying only a few glitches, IBM's computer named 'Watson' easily beat two human competitors on the popular US quiz show Jeopardy!. During the three-day showdown that ended on Wednesday, Watson became a high profile exercise in advanced science and computing, Reuters reports.

The supercomputer, named after former International Business Machines Corp. president Thomas Watson, showed off its encyclopedic knowledge of topics ranging from ancient languages to fashion design. "Vedic, dating back at least 4,000 years, is the earliest dialect of this classical language of India," was one of the clues given by host Alex Trebek.

"What is Sanskrit?" Watson answered in the show's question-as-an-answer format, before going on to solve clues ranging from agricultural policy in the European Union to the designer Marc Jacobs.

One of its few mistakes, however, came up when the computer answered with "Dorothy Parker" instead of "The Elements of Style" and repeating other contestants' mistakes.

In the end, however, Watson defeated the brainy human competitors with $77,147 in winnings while Ken Jennings, who won 74 games in a row during the show's 2004-2005 season, came in second with $24,000. Brad Rutter, who has in previous appearances won a total of $3.3 million, followed with $21,600. IBM plans to donate all of Watson's winnings to charity.

Watson has been compared to Deep Blue, IBM's chess-playing supercomputer that beat world champion Garry Kasparov in 1997, but modern programming and increased processor speed allows it to find answers from ambiguous clues, such as this one: "It's a poor workman who blames these."

"What are tools?" answered Watson.

Turning 100 years old this year, IBM spends around $6 billion a year on research and development even as companies such as Google Inc. and Apple Inc. have become the tech industry's darlings. An unspecified part of that development goes to what it calls "grand challenges," or big, multiyear science projects such as Watson and Deep Blue.

Paul Saffo, a longtime Silicon Valley forecaster, among others, sees better search engines as the ultimate benefit from the "Jeopardy!"-playing computer. "We are headed toward a world where you are going to have a conversation with a machine," Saffo tells AP. "Within five to 10 years, we'll look back and roll our eyes at the idea that search queries were a string of answers and not conversations," he says.

"I see human intelligence consuming machine intelligence, not the other way around," David Ferrucci, IBM's lead researcher on Watson, said in an interview Wednesday with AP.

"Humans are a different sort of intelligence. Our intelligence is so interconnected. The brain is so incredibly interconnected with itself, so interconnected with all the cells in our body, and has co-evolved with language and society and everything around it." The beneficiaries, Ferrucci concluded, could include technical support centers, hospitals, hedge funds or other businesses that need to make lots of decisions that rely on lots of data.


On the Net: