Watson Ties For First Place in First Round of Jeopardy! Challenge

By   /  February 14, 2011  /  No Comments

The team at IBM responsible for Watson must be feeling pretty good about its work right now – and rightly so. On the first day of the Jeopardy! The IBM Challenge, Watson tied for first place with Brad Rutter, each having earned $5,000. Ken Jennings was on the board, with $2,000.

But early on in the game it was no sure thing that either of the former Jeopardy! Champions were going to make a strong mark against the closest thing we have to HAL 9000. While Rutter got the first win for the alternate four-letter word for vantage point or belief (what is view), Watson beat both its opponents to the buzzer the next time around.

And right after that it hit the Daily Double – winning even though it had only a 71 percent confidence in its answer (who is Hyde, the category being Literary APB and the answer noting that we were looking for the murderer of Sir Danvers Carew). From there, up until about the first commercial break, our man Watson was regularly making a killing, its confidence levels in its question-answers regularly hitting the 90-percent plus mark. From this vantage point, Jennings and Rutter were both looking a little stunned.

Highlights and Trouble Spots

Watson lived up to its reputation for being a bear to beat on the more straightforward statements, or those where the factual connections made in the answer seemed more or less along the dotted lines  – “What is The Last Judgment,” “Who is Grendel?” “Who is Lady Madonna?” Where it seemed to me that Watson had the most trouble was when it came to the Decades category, where the revealed answers had little internal contextual affiliations. There really were no discernible connections to the parts of the answers aside from, well, the decade of the question itself. Baseball and Oreos? Klaus Barbie and the discovery of DNA?

As cool as it was to see the queries Watson felt confident to answer, it was equally interesting to see those where its confidence level was lacking. For instance, it had only a 37 percent confidence in the question whose answer was the person who cannot be named but was responsible for killing Mad Eye Moody and Snape. But it didn’t buzz in on Lord Voldemort. Here I have to ask: Was Watson right to be in doubt? Technically wasn’t it Nagini, Voldemort’s snake, who did the actual killing of Snape?

The Twitter feed was, of course, alive with commentary on the action. There was a general feeling that what we were collectively witnessing was pretty incredible stuff. Mixed in, though, were also those comments about how humanity should just give up the ship now, and why does the world insist on trying to create a Terminator?

When I commented to my family that such comments seemed silly – no one was trying to substitute Watson for a human, though he did have such a cute “happy” rays avatar – I was brought up short by a comment from my husband. “You know,” he said, “you’ve been referring to that computer as a ‘he’ the whole time.”

Oh, right.

If you want to be prepared for tonight’s episode with some more background on Watson, have a look:

About the author

Jennifer Zaino is a New York-based freelance writer specializing in business and technology journalism. She has been an executive editor at leading technology publications, including InformationWeek, where she spearheaded an award-winning news section, and Network Computing, where she helped develop online content strategies including review exclusives and analyst reports. Her freelance credentials include being a regular contributor of original content to The Semantic Web Blog; acting as a contributing writer to RFID Journal; and serving as executive editor at the Smart Architect Smart Enterprise Exchange group. Her work also has appeared in publications and on web sites including EdTech (K-12 and Higher Ed), Ingram Micro Channel Advisor, The CMO Site, and Federal Computer Week.

You might also like...

machine learning 300 x 224

Progress in Cognitive Computing and Machine Learning

Read More →