TED Talk: Jennings v Watson

Deep Blue versus Kasparov was an pivotal moment for me. At the time I was playing lots of chess on the computer. (And usually losing.) So the prospect the best player in the world cannot beat a computer was a depressing prospect.  Maybe it is a sign of how much I had matured (or immatured) to think a computer could beat someone better than me at trivia. Maybe it has something to do with routinely using computers to compile things for me in minutes that would take me days.

If you have not seen anything about IBM’s Watson, then start with NOVA’s Smartest Machine on Earth. Even better, here is Day 1 of Jeopardy.

If the below video does not play, then try Ken Jennings: Watson, Jeopardy and me, the obsolete know-it-all.

m4s0n501

Tags: , , , , , , , , ,

  • http://www.facebook.com/jmsearcy Joseph M. Searcy

    I have to say after meeting some of the team that created Watson, and actually getting a chance to compete against Watson, it’s an extremely impressive effort. From the core design of the cluster of 90 IBM Power 750’s to research involved in building the data set, to the analytics behind deciphering natural language to the debugging tools that allowed researchers to visualize Watson’s thought process. What’s even more impressive, and most people don’t know, is that Watson wasn’t searching the web (it had no I teensy access), or analyzing terabytes of data on disks…..all information was found within 512GB of data stored in RAM for the competition. Prior to the competition analytics were done across every episode of Jeopardy to date and statistics were used to limit the data set used to the most frequent categories chosen. That’s just the tip of the iceberg too! Lots of cool things to come from this tech. I’ll also note that Jennings, along with a ha Ful of others have beaten Watson ingesting, but all that matters is the one time that was televised ;)

%d bloggers like this: