|
|
#1
|
|||
|
|||
Is Anybody Watching Watson on Jeopardy?
The tournament on Jeopardy which pairs Ken Jennings and Brad Rutter against IBM computer Watson has been pretty interesting, with the final round to be played tonight. It's clear that the computer has a huge advantage, and it is because it simply can signal in faster than a human. If there is a situation where all three know the answer, I think Watson will get it every time. It takes a human a few milliseconds to bend the thumb and press the buzzer, while a computer signals in electronically. And that is a winning advantage. And of course its response last night in Final Jeopardy was a mystery.
Last edited by barrysloate; 02-16-2011 at 11:20 AM. |
#2
|
||||
|
||||
I agree that the computer has the info proccessing advantage. I'm a Jeopadry junkie and once had a best of 48 out of the 61 answers. It is very entertaining to watch and Watson certainly is an outstanding machine.
My personal nightmare is getting onto the show and having catagories like "Women's fashions","Polo terms","Poetry","hip-Hop music","Things that go through a cat's mind" Rawn
__________________
Not a forensic examiner, nor a veterinarian, but I know a horse's behind from a long ways away. |
#3
|
||||
|
||||
I saw that last night and thought it was pretty amazing. As an owner of IBM, I was pulling for Watson. More impressive than the signal speed, I thought, was Watson's ability to process the most likely top three responses along with the percentage possibility of them being accurate - in less than a second
Was last night the first round or the second? I hope the folks at Jeopardy are going to try to trip up Watson tonight, I was feeling sorry for those poor humans. |
#4
|
|||
|
|||
Rawn- I know what you are saying. In the first round there was a category Beatles Names and I knew all five easily. But if the category were Hip Hop I wouldn't be able to answer even the most basic question. Everybody has topics they know nothing about (forget opera too, I'm a goner).
Tsaiko- it took them Monday and Tuesday to play a full game, as they spent a good deal of time giving the audience a behind the scenes look at the IBM facility, as well as the computer itself. Tonight is the last night and I'm guessing they will play a regular game. Yet I found it amazing that in a Final Jeopardy of U.S. Cities, the computer responded Toronto. It would be most interesting to see how it derailed on it. Certainly Watson has an incredible amount of information that it can process at the speed of light, but its ability to understand the questions, when words often have multiple meanings, is most impressive to me. Last edited by barrysloate; 02-16-2011 at 12:29 PM. |
#5
|
||||
|
||||
Toronto really threw me. The wording puzzles are the hardest part to figure out and not the actual answers.
Rawn
__________________
Not a forensic examiner, nor a veterinarian, but I know a horse's behind from a long ways away. |
#6
|
||||
|
||||
Wow I haven't seen an episode for a decade or so, is Alex Trebec still the host ?
|
#7
|
|||
|
|||
Yes, Trebek is still the host. He's now in his early seventies.
|
#8
|
||||
|
||||
I'm sorry I missed that, what a great PR move for IBM.
|
#9
|
||||
|
||||
my big problem with this tournament was that since watson didnt interpret actual speech, the system fed him a txt file with that question as soon as alex stated reading the question - this was unfair to the humans as with the power of watson im sure it had calculated its answer before trebek was even half way done with reading the question - im pretty sure they knew this as well b/c when they were doing the behind the scenes stuff they quickly glossed over the synchronization issues and just stated how the computer got the questions
|
#10
|
|||
|
|||
I agree that Watson had an unfair advantage with its ability to ring in before the other players, when all three knew the answers. Among three live players, one would expect that each would be successful in ringing in for approximately 1/3 of the clues in such circumstances. Yet, Watson rang in first virtually every time. Maybe a fairer test would be to remove, or at least reduce, the element of speed.
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
WTB 1924 V122 Willard's Watson | Archive | Everything Else, Football, Non-Sports etc.. B/S/T | 0 | 08-25-2008 01:28 PM |
Watching...Bidding | Archive | Net54baseball Vintage (WWII & Older) Baseball Cards & New Member Introductions | 30 | 04-17-2008 01:30 PM |
Whale Watching | Archive | Net54baseball Vintage (WWII & Older) Baseball Cards & New Member Introductions | 3 | 03-24-2008 11:05 AM |
Anybody know eBayer George Watson herman060863 | Archive | Net54baseball Vintage (WWII & Older) Baseball Cards & New Member Introductions | 0 | 12-16-2007 04:29 PM |
F/S: 1933 Goudey Watson Clark SGC 30 | Archive | 1920 to 1949 Baseball cards- B/S/T | 0 | 12-31-2005 12:02 PM |