3 ms·

Watson goes to college: How world’s smartest PC will revolutionize AI

19 points by iProject 14y ago

neop 14y ago
I don't know much about AI, but I highly doubt Watson will revolutionize anything. Significant advances in AI will most likely come from better software and algorithms and while IBM is great at doing hardware, they are not so good when it comes to software. The way I see it, Watson is just IBM doing PR work to get people to know their brand since they don't really offer any consumer products anymore.
- jechen 14y ago
  I wouldn't discount IBM so quickly- to my understanding IBM's been partnering with a lot of institutions to beef up Watson. I know Carnegie Mellon was quite involved through the whole Jeopardy showcase (I even got to watch him school contestants from local universities in Pittsburgh). What's to say better software and algorithms won't be developed as a result of IBM's push to find new use cases for Watson? The article goes into great detail about how Watson has already contributed to the current AI landscape. I for one love that IBM's relocating Watson to a university where students will get to interface with the machine. Haters gonna hate.
- MrHappyOne 14y ago
  I am currently working on a doctorate in ML. I highly disagree with your perspective on Watson. Watson's hardware was not revolutionary; it's software was. Most super computers have far better specs than Watson. Watson's deepqa algorithms are currently the best in the world. These algorithms can be applied to many industries (like medical diagnosis). IBM has been around for a long time, and it is doing fine financially. IBM offers business products; if you have a data center, you're most likely own IBM servers. IBM also offers awesome software solutions for businesses. I am extremely pleased IBM used Watson for PR promotions; it showed the general public the current capabilities of AI.
slacka 14y ago
Watson is a great example of the graduation evolution of weak AI. It succeeds beautifully in its narrowly defined goals, but is still just a glorified statistical search engine. It will not revolutionize AI anymore than Deep Blue did with Chess or the speech recognition software of the 90s did. To quote Jeff Halwkings "As Searle showed with the Chinese Room, behavioral equivalence is not enough. Since intelligence is an internal property of a brain, we have to look inside the brain to understand what intelligence is. As we'll soon see, there is an underlying elegance of great power, one that surpasses our best computers, waiting to be extracted from these neural circuits. ... For half a century we've been bringing the full force of our species' considerable cleverness to trying to program intelligence into computers. In the process we've come up with word processors, databases, video games, the Internet, mobile phones, and convincing computer-animated dinosaurs. But intelligent machines still aren't anywhere in the picture. To succeed, we will need to crib heavily from nature's engine of intelligence, the neocortex. We have to extract intelligence from within the brain. No other road will get us there. " I'd put my money on Deep Neural Networks or Hawkings' Hierarchical Temporal Memory approach to "revolutionize AI".
- dbaupp 14y ago
  Searle didn't show anything of the sort. There have been many responses to it, such as pointing out that it is the whole system, rather than the man, that understands Chinese, or that it is begging the question in terms of the uniqueness of human intelligence. E.g. a fun response which reverses it to be AI talking about humans http://lesswrong.com/lw/ghj/searles_cobol_room/ http://lesswrong.com/lw/ghj/searles_cobol_room/
  - s_baby 14y ago
    His thought experiments give a good illustration of the "hard problem" of consciousness. A) It's not rationally possible to explain qualia in terms of quantitative phenomena. We can draw correlations using empiricism but they are ontologies fundamentally inconsistent with one another. B) Intelligence does not imply experience. One of the biggest problems here is conflation of the two. An amoeba has intelligence with basic consciousness of its surroundings. That does not mean it has experiences.
dave_sullivan 14y ago
There are going to be some revolutions in AI in coming years, but I don't think they will come from IBM (or google or msft or any other similarly established company) In the history of major technological advances, when has a big established player gotten it right? IBM invented the relational database, but Oracle (a tiny startup at the time) still ate their lunch for years. Why? Because IBM had already invested too much in non-relational databases and would have been cannibalizing their own business. Watch the exact same thing happen to Watson, the software powering Google's self driving cars, whatever Kurzweil is working on, etc. Not to mention, many of the current AI luminaries heading research at these companies are near retirement age--not to be ageist, but Einstein did not develop the theory of relativity near retirement, he did it when he was 26. Why would the biggest advances in AI be different? The company that really gets AI right--the company that can call itself "the AI company" like google can call itself "the search company" and facebook can call itself "the social company"--you probably haven't heard of it yet.
gojomo 14y ago
Watson's strongest and most unique advantage may be IBM's PR department.
treerex 14y ago
"PC"? Watson's hardware weighs in at around 40 tons total.

19 points by iProject 14y ago

neop 14y ago

I don't know much about AI, but I highly doubt Watson will revolutionize anything. Significant advances in AI will most likely come from better software and algorithms and while IBM is great at doing hardware, they are not so good when it comes to software. The way I see it, Watson is just IBM doing PR work to get people to know their brand since they don't really offer any consumer products anymore.

jechen 14y ago
I wouldn't discount IBM so quickly- to my understanding IBM's been partnering with a lot of institutions to beef up Watson. I know Carnegie Mellon was quite involved through the whole Jeopardy showcase (I even got to watch him school contestants from local universities in Pittsburgh). What's to say better software and algorithms won't be developed as a result of IBM's push to find new use cases for Watson? The article goes into great detail about how Watson has already contributed to the current AI landscape. I for one love that IBM's relocating Watson to a university where students will get to interface with the machine. Haters gonna hate.
MrHappyOne 14y ago
I am currently working on a doctorate in ML. I highly disagree with your perspective on Watson. Watson's hardware was not revolutionary; it's software was. Most super computers have far better specs than Watson. Watson's deepqa algorithms are currently the best in the world. These algorithms can be applied to many industries (like medical diagnosis). IBM has been around for a long time, and it is doing fine financially. IBM offers business products; if you have a data center, you're most likely own IBM servers. IBM also offers awesome software solutions for businesses. I am extremely pleased IBM used Watson for PR promotions; it showed the general public the current capabilities of AI.

slacka 14y ago

Watson is a great example of the graduation evolution of weak AI. It succeeds beautifully in its narrowly defined goals, but is still just a glorified statistical search engine. It will not revolutionize AI anymore than Deep Blue did with Chess or the speech recognition software of the 90s did. To quote Jeff Halwkings "As Searle showed with the Chinese Room, behavioral equivalence is not enough. Since intelligence is an internal property of a brain, we have to look inside the brain to understand what intelligence is. As we'll soon see, there is an underlying elegance of great power, one that surpasses our best computers, waiting to be extracted from these neural circuits. ... For half a century we've been bringing the full force of our species' considerable cleverness to trying to program intelligence into computers. In the process we've come up with word processors, databases, video games, the Internet, mobile phones, and convincing computer-animated dinosaurs. But intelligent machines still aren't anywhere in the picture. To succeed, we will need to crib heavily from nature's engine of intelligence, the neocortex. We have to extract intelligence from within the brain. No other road will get us there. " I'd put my money on Deep Neural Networks or Hawkings' Hierarchical Temporal Memory approach to "revolutionize AI".

dbaupp 14y ago
Searle didn't show anything of the sort. There have been many responses to it, such as pointing out that it is the whole system, rather than the man, that understands Chinese, or that it is begging the question in terms of the uniqueness of human intelligence. E.g. a fun response which reverses it to be AI talking about humans http://lesswrong.com/lw/ghj/searles_cobol_room/ http://lesswrong.com/lw/ghj/searles_cobol_room/
- s_baby 14y ago
  His thought experiments give a good illustration of the "hard problem" of consciousness. A) It's not rationally possible to explain qualia in terms of quantitative phenomena. We can draw correlations using empiricism but they are ontologies fundamentally inconsistent with one another. B) Intelligence does not imply experience. One of the biggest problems here is conflation of the two. An amoeba has intelligence with basic consciousness of its surroundings. That does not mean it has experiences.

dave_sullivan 14y ago

There are going to be some revolutions in AI in coming years, but I don't think they will come from IBM (or google or msft or any other similarly established company) In the history of major technological advances, when has a big established player gotten it right? IBM invented the relational database, but Oracle (a tiny startup at the time) still ate their lunch for years. Why? Because IBM had already invested too much in non-relational databases and would have been cannibalizing their own business. Watch the exact same thing happen to Watson, the software powering Google's self driving cars, whatever Kurzweil is working on, etc. Not to mention, many of the current AI luminaries heading research at these companies are near retirement age--not to be ageist, but Einstein did not develop the theory of relativity near retirement, he did it when he was 26. Why would the biggest advances in AI be different? The company that really gets AI right--the company that can call itself "the AI company" like google can call itself "the search company" and facebook can call itself "the social company"--you probably haven't heard of it yet.

gojomo 14y ago

Watson's strongest and most unique advantage may be IBM's PR department.

treerex 14y ago

"PC"? Watson's hardware weighs in at around 40 tons total.