Slashdot Mirror
Watch IBM's Watson On Jeopardy Tonight
JohnMurtari notes that the media hype machines are massively promoting tonight's battle between Jeopardy champions and a super computer. Yes it's a PR stunt. But I imagine the actual research probably had a lot of interesting problems to address. Anyway, you can learn about IBM Watson if you're interested. I'm sure the most amusing bits will be on YouTube about 30 seconds after air time.
JohnMurtari notes that the media hype machines are massively promoting tonight's battle between Jeopardy champions and a super computer.
I'm so glad we're above that.
Seriously, if this thing doesn't accidentally observe the Higgs Boson while seeking for a question to an answer, I'm going to be disappointed.
My work here is dung.
I mean its going to be one of the first times that a robot with speech recognition will be live and responding against people in real time on broadcast TV. I think you all have been living in your movie plots too much to realize how big of a moment this actually is.
Furious over losing to fleshy things, shortly after the show, Watson renamed itself SkyNet, created a Cylon girlfriend for itself, and set about to eliminating all fleshy things.
Truth, Just Us, And Hatred For All Mankind!
It's a little more interesting than a computer beating a human at chess, which is completely algorithmic. However, at its heart, this is simply an exercise in in data storage, lookup, and statistical probabilities in determining a likely answer. It does not involve any artificial intelligence or machine intelligence at all. From a purely technological standpoint, it's quite impressive what IBM has been able to do. It'll be even more impressive in 10 years when the same type of power is in my phone.
http://video.pbs.org/video/1786674622/ for Americans. :)
Ant(Dude) @ Quality Foraged Links (AQFL.net) & The Ant Farm (antfarm.ma.cx / antfarm.home.dhs.org).
There was an interesting episode of Nova called "Smartest Machine on Earth" that was pretty interesting. It talked a lot about the challenges they faced, how they addressed them, what adjustments they made along the way, etc. I don't see the episode listed on the schedule for replay any time soon, but you can watch it on the website
http://www.pbs.org/wgbh/nova/tech/smartest-machine-on-earth.html
What a wishy-washy summary. It's not like you have anything better to do tonight than watch Jeopardy, is it?
Speaking of which, it seems like *I* was supposed to buy or do something tonight... now what was it...
Adidas To Bring Back Sneakernet
Yes, they are hyping it for more than it's worth, but there's one thing here worth hyping: the fact that IBM still runs a state-of-the-art computer science research program. Just about every other corporation has gutted whatever research institute they had and concentrated their R&D funds on directly marketable products. IBM still runs the Watson research center that develops ideas from basic computer science down to products, even if it takes more than a decade to do it and the results are not certain. That is certainly worth respect.
A really fun question to ask the IBM sales people would be: "If I ran this on one of your mainframes, roughly how much would it cost me in MIPS fees?". I can't believe they still get away with that.
IBM's move into online SEARCH. This is going to be interesting.
Yours In Novosibirsk,
Kilgore Trout, C.I.O.
While obviously it does utilize data storage, lookup, and statistical probabilities to help find and choose an answer, it also significantly relies on machine learning (a large branch of AI) to understand the questions and choose the answers.
Popisms.com - Connecting pop culture
I love the schizophrenic nature of the post. Every other sentence is a reason you shouldn't watch Jeopardy. But the poster clearly wants to watch it himself.
Basically if you like Jeopardy, watch it. It will be good. Ken Jennings will be back. It' not just going to be a computer, it's going to be two really good people playing.
Even if you don't particularly like Jeopardy, you might still like to watch it, if you want to see an unusual spectacle, or are interested in something vaguely related to artificial intelligence. And if you don't want to watch it, then don't, it's just entertainment after all. But don't whine about it, that's even worse.
My guess: the computer wins, not because of its massive database, but because it can push the button really really fast on the questions it does know.
"First they came for the slanderers and i said nothing."
Having seen a video of it in action, I'm very impressed.
http://www.engadget.com/2011/01/13/ibms-watson-supercomputer-destroys-all-humans-in-jeopardy-pract/
Where can I find streaming video of the episodes online? I don't have a TV, but since I'm in the field I want to watch more than just the embarrassing outtakes.
They actually showed the beginning of the filming of tonight's show.
The first time the Jeopardy producers saw Watson in action, the performance was erratic. But a fairly simple change made a respectable improvement. That was to use the responses from the other players and the host. This feedback reduces ambiguity for later answers. The improvement was enough to make the producers use Watson then.
True, but it's also an understanding of human language. If you watch the PBS NOVA episode on it, it can be quite hard. Like the category called "Days in months", where you're given two days of a month and have to answer in the month. How does a computer figure that out? (In Watson's case, it didn't until it saw the correct answers and figured out that it needed to be months).
Or a category like "before and after"?
Pure trivia questions - yes it's a simple database lookup (and Watson basically kills at it). But Jeopardy isn't just a nerd trivia game, it's all about subtleties of language - double meanings, puns, wordplay and other elements that make it extremely hard.
It's basically a step towards understanding natural language, with all the issues and subtleties that we put in - emotions, sarcasm, etc.
Or, in Feb 14-16, 2011, Skynet will show off its ability to understand human language.
Lots of people think computers are powerful enough to talk and play games. They may not understand Watson merely uses search and inference without much understanding of language. It could do just as well in Chinese.
I'm torn. I actually want to like this spammy post. it's just so cute sitting here at the bottom of the thread. And it's so polite.
From what I've seen, its more like 19/20 nerd trivia, 1/20 subtleties of language. I would expect it to do very well at 95% of the type of questions generally asked. It's all about what the jeopardy authors choose as categories which will make or break its game.
It does not involve any artificial intelligence or machine intelligence at all
You think saying that makes it true? At this point we have machines that can read, hear, reason and plan. Fly a plane, drive a car in traffic ... it isn't intelligence they're lacking, it's desire.
As always, all IMO. Insert "I think" everywhere grammatically possible.
>> While obviously it does utilize data storage, lookup, and statistical probabilities to help find
>> and choose an answer, it also significantly relies on machine learning [wikipedia.org] (a large branch of AI)
>> to understand the questions and choose the answers.
If this machine chooses the Answers it will lose at Jeopardy, It needs to determine the Questions . . .
We are Dead Stars looking back Up at the Sky
The ability to follow a rigid set of very specific instructions that someone else has established is not intelligence.
"But this one goes to 11!"
The show that will air tonight has been filmed weeks ago, for no good reason that I can think of. They would definitely stir up more attention and attract more viewers if the show was broadcast live. This is like being told months after the fact that Deep Blue did in fact beat Kasparov, and the moves were 1. Kp3, ... That would have been completely lame.
What IBM hopes for is for Watson to win, but not win by much, so that people aren't put off by its brutality. And this taping of the show weeks ahead of the airing just invites speculation that the game was rigged to produce exactly this result. After investing so many resources in Watson, it's pretty dumb of IBM to not do this last thing right - which would have greatly raised the interest without any additional cost. One imagines that they did this because of their lack of confidence in Watson's performance. And that makes them look far less badass than they otherwise would.
AHAHAHAHAHAHA Awesome! +1 Funny if I had mod points!
Congratulations. I estimate that 80% of the human race fails your particular Turing analogue. It actually explains a lot, really.
Welcome to the Panopticon. Used to be a prison, now it's your home.
80% - you sir are a cock-eyed optimist!
"But this one goes to 11!"
Upon winning will Watson cry out, "OH, Yeah! Take that meatbags!"?
A machine that interprets human language and produces spoken answers is showing some of the most important parts of human intelligence, and it does so artificially. So this is very much artificial intelligence.
any standard. Machines for the past decade have mostly been groomed to fight wars, as their human counterparts have so needlessly done. Now, for the first time, machines are being taught to humiliate living rooms and nursing homes full of doddering elderly tv viewers with decisive answers to some of americas most inane questions.
seriously though, the watson project is pretty freakin cool even if its just a PR stunt.
Good people go to bed earlier.
I will watch it to see if Alex can chit-chat with it with regular questions, like "Where are you from?", "What's your job?", "Married? Any kids?", "Do you like... human?"
Jeopardy! is the game where an answer is given and the contestant must supply a question...
Once their AI is sufficiently good at it, they're obviously planning to give it the answer 42.
Jeopardy airs 2/14, 2/15, and 2/16 at 7PM, EST on ABC, for anyone curious. :)
Jeopardy is syndicated. It will show on different networks and at different times depending on where you live.
I for one hope that Watson's text-to-speech engine fails miserably and he starts mispronouncing category names like "The Pen Is Mightier".
It's AI, but it's weak AI. Wake me up when it does something they didn't program it to do. (OK, I'm awake now, I think it's pretty cool, but it's not like they figured out the algorithm the human brain uses to think or anything. At best only part of it).
"First they came for the slanderers and i said nothing."
Wrong. It is absolutely "artificial intelligence." It isn't science fiction AI, it's real AI. Here in the real world, the term AI merely refers to the sorts of problems which human brains can do with ease but synthetic computers can't do. Playing Jeopardy fits firmly within this definition.
Your phone will not have terabytes of RAM in ten years. I promise.
A slashdotter who didn't build his own computer is like a Jedi who didn't build his own lightsaber.
Watson won...easily.
"Here Lies Philip J. Fry, named for his uncle, to carry on his spirit"
It does not involve any artificial intelligence or machine intelligence at all
One beauty of artificial intelligence is that once you have solved an AI problem, it is no longer an AI problem. It's becomes an algorithm, or a database problem, or a statistics problem, etc.
Chess was once considered AI. Heck, tic-tac-toe was once considered AI. Solving equations and integration were once considered AI (and a certain level, still is). Playing Jeopardy is typically considered a measure of human intelligence. So it is funny that having a computer do it is not considered artificial intelligence.
Watson's voice is indentical to that of HAL 9000, I'm not interested. It would be hilarious if 2001 was a topic, though.
Who says our brains work any differently?
In fact it's easy to argue that the random arrangements of nuero-transmitters and nueroreceptors in your head dictate from birth every reaction to every experience you will ever have.
The only thing that sets us apart from animals is a) the ability for abstract thought b) thumbs. If your argument was that computers need to be able to think abstractly about the world and how to combine things to make new things, then maybe you're on to something. If your argument is that humans are somehow not subject to strict hard rules on our behavior, I think you overestimate humanity.
My local station ran round one at 11am, and I've posted spoilers here: http://twitter.com/robotwisdom
It is the first song played by a computer, in 1961.
I think it has to OCR it off of the blue screens, as humans do.
-d
"Here Lies Philip J. Fry, named for his uncle, to carry on his spirit"
They had a special on Nova a few nights ago. Had you watched this, you would know that Watson combines several artificial intelligence and machine learning technologies to achieve it's level of proficiency. However, it still does not "understand" the problems it is solving.
Who says our brains work any differently?
I do. When I don't have the information needed, my brain doesn't stop processing things and put up an "ERROR". I have to somehow come up with that information through indirect discovery, or make a reasonable guess based on the information I do have. Yep, it is abstract thinking, and while some computer programming can effectively fake some of this behavior, it is far from being perfected.
As for your supposition what separates humans from animals - some primates (chimpanzees, orangutans, etc.) do have opposable thumbs, so that can hardly be a category that separates us. Abstract thought has not yet been conclusively proven in the animal world, but is that even possible to prove or disprove?
"But this one goes to 11!"
As if your own intelligence were from some different astral plane.
What did you think intelligence was?
It'll be even more impressive in 10 years when the same type of power is in my phone.
Your phone will not have terabytes of RAM in ten years. I promise.
The actual phone may not have 10 terabytes of RAM, but I bet it will be able to instantly access computers that do.
In fact, with Google, Bing, and other search engines, one could make a case that your phone already does contain at least that amount of RAM.
We are also seeing this with services like OnLive. Once we get to a point with mobile phones that have very high bandwidth, very low latency and access to dedicated, powerfull remote computers, everything is going to go to dumb terminal renderers.
The only problem I have with the way Watson is being used on Jeopardy is that gets a minor head start on being able to answer the question. Although the text of the question is actually revealed to everybody, including Watson, simultaneously, humans have to take a second (or two) to actually read or hear the clue to even know what it is actually saying, whereas it's my understanding that Watson receives the clue electronically, and thus knows all of the text in the clue in mere milliseconds, before the human contestants have even finished reading/hearing it.
It has been my supposition since I first heard of this that Watson would tend to fare poorer against human competitors when the clues are very terse, because then Watson doesn't get as much extra time over the human competitors to parse the clue and search its database for possible answers.
I look forward to finding out if I was right.
File under 'M' for 'Manic ranting'
"Your phone will not have terabytes of RAM in ten years. I promise."
No, but it will have something substantially better; access to a cloud-based "Watson" that is constantly learning from human interaction just like the current Watson.
...It's basically a step towards understanding natural language
In the Nova episode you mention, they make it clear that this is in no way comparable to "understanding". It is still merely numerical manipulation.
Yeah, because the internet needs a search engine that can respond to erudite puns with the appropriate question.
Is the Jeopardy! problem that different from responding to real-world natural language questions with the appropriate Google query?
Uhh...a computer system can keep trying when it hits an error. We programmers put stops in it because we know the computer will screw up once it leaves the reservation of known working states.
Humans keep running...and usually when they do something they don't know what they are doing, something bad happens they are lucky to survive...
What is "Kebert Xela?"
Watch tonigh and prepare to believe a machine can think.
Being able to answer a certain class of questions (ok, determining a correct question from an answer) does not demonstrate self-awareness. I'm not sure that extrospection without introspection qualifies as thinking, at least, not in the human sense.
However, at its heart, this is simply an exercise in in data storage, lookup, and statistical probabilities in determining a likely answer.
What, please, is AI if not exactly that?
This system analyses the questions in the game, expressed in highly idiosyncratic human language. How is that done? Add "modeling" to your list above, and that's it. Data storage, lookup, and statstical probabilities. Turns out that's highly likely the way *we* understand human language, too, with an emphasis on the last part.
The system then searches its knowledge base. How is that done? As described above. How do we, as humans do it? We don't fully understand, but you can bet your bottom dollar that it's an exercise in data storage, lookup, and statistical probabilities in determining a likely answer. There's plenty of research in the neuroscience literature (I am a neuroscientist and an alumnus of MIT LCS/AI) about evidence gathering and decision making, and it would appear that we, or at least animals, are highly predictable in that respect as if we are using a fixed, very simple algorithm.
So, if AI isn't doing what humans and other animals do, and therefore what this program Watson is doing, then please enlighten us: What is AI?
Put my fist through my alarm clock with its ding-dong death inside my ear. - The Blackjacks.
Yes, but it is always called artificial intelligence until some one figures out how to make it work, then everyone starts calling it machine learning.
Well it did kind of do what you are talking about using machine learning.
There was a special on NOVA about it the other night that went pretty in depth about how it works.
For instance, there was a category that had answers like
Martin Luther King Jr. Day, and New Years Day
and at first watson answered "what are Holidays"
But the answer was "what is January" because the clue was something like " days in ... "
And after two of the human players, each answered months for the next three in the category, it finally chimed in on the last answer and correctly said
"what is may"
Is it just me, or does anyone else think this represents a huge threat to Google? Think about a search engine that can understand your question. Talk about refining search results substantially. Watson's major achievement here seems to be a big dent in understanding language.
Lot's of techie folks have DVRs. In fact, I just used my iPhone to instruct my DVR to record these Jeopardy episodes.
A suffciently-functional weak AI is indistinguishable from the real thing.
Don't blame me, I voted for Baltar.
It's still far better than Google and other search tools currently available. Type a jeopardy question into Google, and click "feeling lucky" -- you won't find squat. Jeopardy is all about obscure clues. "This man was the son of a president, a president himself, and invaded the same country as his father." Type that into google and you'll get crap, because "Bush" and "Iraq" are never mentioned in the clues, because they'd be too obvious.
IBM has zero business in the laptop/desktop market. They would be ecstatic with all Mac shops with the current state of Apple not giving a crap about the datacenter compared to all HP or all Dell.
XML is like violence. If it doesn't solve the problem, use more.
What makes you think you're so different, particularly in the domain of how you know trivia? It's not like you have a particularly more concrete grasp of birthdate of John of Gaunt than a machine does.
Don't blame me, I voted for Baltar.
There's not consensus on what artificial intelligence is because there's not a consensus on a precise definition of intelligence. One thing that designing computers and software to do such things as play tic-tac-toe and chess has shown us is that the ability to do such complex logical tasks is very different from human intelligence. It also shows how little we understand of the human brain and mind.
"This man was the son of a president, a president himself, and invaded the same country as his father."
I'm feeling lucky...!
http://en.wikipedia.org/wiki/George_W._Bush
How does Watson push the button? Or more precisely, does it wait until it has determined a good response before it buzzes in? Because half the game is beating someone else to the buzzer when you just think you'll be able to figure out the question in time, not because you already know it.
I've abandoned my search for truth; now I'm just looking for some useful delusions.
'Abort, Retry, Fail?' was the phrase some wormdog scrawled next to the door of the Edit Universe project room. And when the new dataspinners started working, fabricating their worlds on the huge organic comp systems, we'd remind them: if you see this message, always choose 'Retry.'
Bad'l Ron, Wakener, Morgan Polysoft
--Sid Meier's Alpha Centauri
"... at its heart, this is simply an exercise in ..."
Why slashdot sucks in a nutshell: If it is that "simple", why hasn't anyone (particularly YOU or anyone else dismissing this around here) done it yet?
I'm just going on the record as saying this is one of the seminal moments in human history and will be remembered as such in years to come.
The applications for Watson are huge. One "Watson" could handle taking the orders for hundreds if not thousands of McDonald's drive-thrus, perform many types of customer service roles in a phone or chat capacity, run kiosks at Lowe's telling you where items are located or what they're normally used for. There's probably tremendous opportunities in statistical/actuarial fields as well as tax accounting.
This is how the technology derived from Watson is going to make a difference. It's going to march us even further down the path of not human obsolescence, but labor obsolescence, a path humanity has been going down for some time. This is why it's important, because if "Watsons" start doing these jobs, it destroys one of the principal tenants of capitalism. Namely that people barter their useful labor with employers in exchange for pay. As this tenant becomes more and more false, society is going to have to take a long hard look at what role people play in it.
Yes, but only if you program it to keep trying. See what I am getting at here?
"But this one goes to 11!"
It has been my supposition since I first heard of this that Watson would tend to fare poorer against human competitors when the clues are very terse, because then Watson doesn't get as much extra time over the human competitors to parse the clue and search its database for possible answers.
It seems like speed in answering is the least interesting part of all this. Especially when you're talking about the difference between six seconds and ten seconds or whatever. If the difference is between six seconds and six thousand years then we might care, but if it's a matter of replacing the racks with POWER 8 processors and trying again, what does it matter?
The interesting thing is whether it can get the right answer in any reasonable period of time.
"What would be a good definition artificial intelligence that wouldn't be subject to goalpost moving?"
When the AI starts chiming in on where the goalpoast is, and giving us suggestions, I think then we can hang our hats and go home as a dying species.
Mod parent up
People will still barter their useful labor with employers in exchange for pay.
Now, they won't have any useful labor to barter with.
Nobody is going to 'take a long hard look at what role people play in society'. You are going to have a clusterf*** of politicians trying line their cronies pockets while the masses basically drop out of the employment system, move in with relatives, etc. Some of them migrate somewhere else. Some will become homeless, some will die from lack of health care.
A good portion of them might get retraining and new employment. Of course, that takes money a lot of them don't have and can't get.
This is either a brilliant troll or you didn't bother putting any time into researching what you just claimed. Either way, it's right at home at Slashdot.
Rob
I thought that Waston was stand alone and they just fed it a ton of data. They didn't want to give it internet access. Doesn't that tell you something?! ;)
You have to understand the semantic meaning of an error in computing. An error is something that is generated by an implementation upon the failure of a test at some level of the system -- it indicates the system has entered a state where further inputs will no longer map to the "desired" outputs. The issue is in how we define "desired," and we find that this is always defined semantically by the humans designing the system, a priori. A computer cannot divide a number by zero, or dereference a null pointer, because we say so, because we apply that abstract truth to the system. We do so because hardware and software form an entity that requires internal consistency to respond to inputs, and when that internal consistency is lost the system no longer is useable.
Humans make errors all the time, it's just that we do not generally halt when we make them. We have other ways of reconciling errors, things we call "rationalization" or "denial" or "learning." Human beings have very limited a priori desired outputs and exception states, and none of them apply to symbolic reasoning -- a coma might be an example of an exception state, and it's brought about by "recoverable device failures." The human brain and cognitive system is also much more finely engineered and rigorous than a computer system, inputs and outputs are always "sane," the states of the system, such as they are, are highly distributed in time and between functional units, and on most levels of operation the global system cannot lose internal consistency in a way that jeopardizes operation.
Well, the Nova ScienceNow that directly preceded the Watson episode was all about animal cognition (probably not coincidentally), and they had several rather unsettling demonstrations of a dog that could remember dozens of toys by name, and collect novel toys given nothing but the novel toy's name; a parrot that could count to eight and construct declarative phrases of nouns and modifiers; and dolphins with functional vocabularies that were provably communicating with each other through their squeaks to collaborate on a trick that they invented themselves.
Most creature's brains are capable of abstraction to a degree, but the physical attributes that are associated with humanity, like the opposable thumb, bipedal walking, and particularly a voice, have the effect of creating enormous selection pressures upon the brain. A hand grabbing a pole can kill one animal a year or a hundred, depending on how smart the brain behind it wields it. It may take one individual one lifetime to teach one other individual how to make a tool, or in the same time teach ten-thousand, completely depending on how well they use speech. Because birds and dolphins and dogs can't really manipulate their environment to the degree a creature with a hand can, the selection pressures fall upon other parts of their physiognomy.
Don't blame me, I voted for Baltar.
At IBM "Lotusphere" event, the closing general session was a preview match. It was done as close the the same was as an actual jeopardy show but with contestants picked from a small mini-tournament of attendees, and a comedian as host instead of Alex T.
Having been present at this (and getting my picture taken next to the Watson "icon/screen") and then watching the Nova episode, I can say for sure that the Nova show was a very well done description of what happens; as well as Watson's strengths and weaknesses.
I'm not sure if they'll show it on the live TV show taping, but in the run through we saw, they showed Watson's top 3 picks with a level of confidence on each. It was as interesting to see the second and third choices as it was to see what it actually came up with for an answer.
A couple of things were updated from when they must have taped the Nova show. First, Watson was far more strategic when it came time to place bets than it had been shown on Nova. Second, it was far better at understanding weird language in the categories.
I'm looking forward to the show.
The problem with quotes on the internet, is that nobody bothers to check their veracity. -- Abraham Lincoln
I didn't see him put forth any Turing analogue. He only claimed that following a very specific set of instructions established by someone else is not intelligence (i.e., it is *not* a valid Turing test) [and he's right].
I said opposable thumbs in my original post, but you're definately correct. Upright walking - in combination with the thumb - is the real winner. Knuckle walking members of the ape family (all of them but humans) don't have a freed up hand to use the tools they make while walking.
So long as computers continue to be built as they are, they will never be anything but numerical manipulation. At their fundamental level, that is all they do. They manipulate numbers in various ways. At the fundamental level there is just binary data, without type or form. In memory the data isn't even characters or pixels or any of that, just a long string of binary digits. It is only given form by the programs that are written to use it.
However that doesn't mean that the behaviour at the higher level will not emerge to be like that of humans to be able to understand, as in correctly parse and work with, human speech and concepts.
After all at a fundamental level humans are just neurons making electrical and chemical signals. When you get down and look at a single neuron and how it functions in the brain it gives no clues that an intelligence might emerge from it.
So in the end, computers will always just manipulate numbers, unless we change how we build processors. That doesn't mean that their numerical calculation may not give rise to something that can accurately be called "intelligence".
Most importantly - human intelligence which itself is emergent, essentially "at its heart, ... is simply an exercise in in data storage, lookup, and statistical probabilities in determining a likely answer"
... approaching this in "intelligently accurate" way would be a major waste of brain tissue for most of our evolution and not helping much, anyway.
... but "intelligent"? While our abstract thought is a very recent, experimental, unoptimized hack. I doubt if that much different in experience from many higher animals - except that we got ourselves a quite complex way of communicating with ourselves, internally (talking to ourselves is what we usually call "thinking")... but it's not hard to notice that this is not our only mode of operation! Those other are IMHO when we experience the "animal levels" of consciousness. Probably not that different, all things considered.
;) )
Diffusion MRI/CAT scan images (revealing directional bias of the fibres in the white matter, the long-distance communication) show a nice "tree"-like network... which of it is intelligent?
One won't find it in a neuron, or in (certainly at least) most areas of the brain used for separate specific purposes (say, name of the object vs. its orientation, distance or color - taking separate paths from the image or sound input)... which are yet crucial for the idea of "ourselves". Stroke (brain, generally) clinics have tests which evaluate damage on this basis - it's possible to tell what an object is, but not its orientation or distance; or knowing there's is an object (and where it is), but not being able to name it (but being able to point to something of even just similar category). There's even one very localized brain trauma which makes people blind... without them realizing it! Couple that with how split-brain patients seem almost normal, and you have an idea how much of a "grip" we have on ourselves.
We are good at "cheating" the complexity of problems requiring "intelligence" - not only distributing workload to (fairly unintelligent by themselves) portions of the brain, also living in approximations, a lot of best effort guessing, settling on quite likely result and taking on another task; focused on rapidly getting sloppy answers (suitable for a sloppy world, with sloppy inputs) - quickly identifying quantities works in the style of "none, 1, 2, 3, 4, 5, 6 or 7, few, a dozen or so (then 2 or 3 of them), oh gods that's a lot"
Running mostly in the opposite direction to a predator, with pseudo-random evasive actions and last ditch efforts, is plenty good enough
That said, dragonflies display a spectacular (and similar in style to xkcd above... only in 3D) hunting technique
However - confidence in our abilities, oversensitive alertness and internalizing experiences mostly of fellow humans were certainly also very useful evolutionary traits. Right there is probably large part of our "absolutely unique intelligence" (or, say, convincing ourselves how good our memory is, how we are a mostly unbroken individual consciousness - while being generally closer to our peers than to ourselves at some different life stages; while not being able to recall much about our experiences or thoughts from the first week of March... 1996)
Ask few random fellas how long it takes for the Earth to go around the Sun (a test from one of Carl Sagan's books, and the sad example was about some students of top uni). Yes, we might wonder how well those with sad answers paid full attention to the question, or if they answered without listening (another thing to explore)... well, that is also the point!
(love the nick, BTW
One that hath name thou can not otter
http://science.slashdot.org/comments.pl?sid=1994992&cid=35200456 You were three minutes late. :) I win! [grin]
Ant(Dude) @ Quality Foraged Links (AQFL.net) & The Ant Farm (antfarm.ma.cx / antfarm.home.dhs.org).
On the Nova episode, I believe they said they "texted" the answer to Watson. (I presume they really mean they somehow transfer it, and not actually use SMS. I don't see a transcript of the episode online to confirm.)
However, while you also mentioned reading, I don't think it's as much of an advantage. We can always read the answer much quicker than Alex can speak it.
IIRC, when there was a blind player, the only extra he got was a Braille list of the categories, not each answer in Braille...
In the future, some guy will be sitting around with his best buddy watching watching football while his personal robot does his housework, balances his budget, finishes his homework, and organizes his love life. The robot will pause to tell a joke it just thought up spontaneously. After he ROFL, the guy will say to his friend "Dude, my new SuperRobot 2400+ is so smart!"
His friend, a computer programmer, will say "That isn't artificial intelligence. It's just a computer program."
Think of it as someone asking Watson "Who is Nerdfest", and you can guess how many will pay for the system.
Using all the Internet logs, I'm sure Watson could identify just about anyone on Earth.
I personally think that this is the exact reason why watson is pretty amazing. We get really caught up in WORDS... but words themselves are only representations of IDEAS.
That's the interesting thing about Jeopardy... it often phrases "answers" in such a way as the words themselves don't matter, it's the ideas behind the "questions" that really matter. When watson goes through it's numerical manipulation of the words... it seems that the computer is trying to numerical understand concepts and ideas and not just words themselves.
I watched the NOVA special on IBM's Jeopardy-playing machine. Essentially, it's a huge search engine coupled with algorithms for choosing answers. It is able to answer 90% of the questions asked, which looks impressive at first. The 10% of wrong answers it gives, however, are pretty bloody stupid. Obviously, there is really no reasoning involved here. Give me access to a massive search engine, and a computer-fast method of accessing it, and I will get 90% of the Jeopardy answers right too. Probably more than that.
Which is not to say the project is a complete waste of time. They've learned a lot by building and tweaking Watson. It's just not artificial intelligence.
Proverbs 21:19
You forgot to insert the part about "UNMARRIED men... get laid." The rest of us poor schmucks thought we were getting a good deal when we bought the cow.
It's not like you have a particularly more concrete grasp of birthdate of John of Gaunt than a machine does.
No. Quite the opposite. The reason I know I'm different in the way I grasp the birth date of John of Gaunt is because I can generalize the concept of birth date in ways that a machine cannot. Your example is a perfect one to demonstrate my point. I can look up on Google and find out that the man known as John of Gaunt was born March 6th, 1340. That seems like a reasonably well established factoid. However, is this the birth date of John of Gaunt? I subject to you that it is not. It may well be the birth date of the person who came to be known as John of Gaunt, but I'm reasonably certain that he was not known by that name on that date. It would have been many years later that he would come to be known as John of Gaunt. So I would argue that this is the birth date of John of Gaunt.
Would a machine understand this distinction? Could a machine even parse this distinction?
I can consider the birth date of other things that a machine would not comprehend - the birth date of the universe; or the birth date of my car; or the birth date of manned space flight. There are an infinite number of things I can consider the birth date of. Things that were neither born nor have a specific date to associate.
I can also consider of the birth date of an idea. I can even consider the birth date of the specific idea that I can consider the birth date of an idea.
While it may be possible to train a machine to make similar generalizations, it is not possible to train a machine to generalize any arbitrary concept. That's how I know there's something different going on in my brain.
I watched that show. The prior trials of the system went poorly. The machine learning really kicked in when (just like a real player) Watson got to"hear" (via text message) the other players answers. And that is when it could make the leap in understanding about the nature of the category.
BTW, the video is on PBS:
http://www.pbs.org/wgbh/nova/tech/smartest-machine-on-earth.html
You have the right to remain sentient. If you give up the right to remain sentient, you will be elected to public office
Jeopardy players aren't permitted to ring in until Alex has finished reading the answer. Actually, the buzzers aren't activated until about half a second after Alex finishes reading. When you can buzz in is signaled by a light that comes on that is not seen on TV, but can be seen by the players. If you buzz in too early, your buzzer is locked out for some period of time (one quarter of a second IIRC).
The best Jeopardy players (and Brad Rutter and Ken Jennings are certainly among the best) will quickly scan the written answer while Alex reads and, if they're reasonably certain they can provide the question, buzz in reflexively as soon as that light comes on. Actually, to be more accurate, often times they'll start the physical motion of buzzing in milliseconds before the light comes on, anticipating that their finger will depress the button right when that light illuminates.
It's likely that Watson's ability to detect when it's okay to ring in, and then quickly do so, is probably superior to most human reflexes. However, that advantage is not related to reading speed when up against the best human players, and is probably measured in milliseconds.
Thank you. You summed up my thoughts nicely.
I've refrained from commenting on this topic because every time the topic of AI comes up on /., the same group of folks drag out the "AI hasn't progressed at all in over 50 years" crap and I get tired of aguing against them.
Their argument boils down to this. If programmers have to program it, then it's not AI. WTF? By that definition I guess all we can do is keep shovelling sand into a fire and hope that a chip falls out that can produce the Unified Theory with no human intervention. Anything less is "just a bunch of algorithms" and not intelligence.
Debating about what constitutes intelligence (artificial or not) with people that take that position will only give you a headache.
Sometimes the light at the end of the tunnel is the headlight of an oncoming train.
I can't wait until Watson takes on Sean Connery, Burt Reynolds and Marilu Henner.
Here's a link to the Nova coverage.
As an adult, with years of nonstop data input you are able to do this. With literally years of nonstop data input, who knows what machine learning could yield. Remember, children are incapable of abstract thought up until at least age 2, and in some cases, much later. There are well defined tiers of mental development describing what growing humans can and cannot do. Its not that the machine is anything different from you, if anything, its starved of data.
I'm going to warm up for tonight's broadcast by watching Maury.
Nothing like prefixing a little artificial intelligence with some genuine stupidity.
I have a Chinese Room that wants to talk to you.
4096R/EF7BAFA6 79E1 DF98 D09D 898F 9A11 F6F0 DDDC 23FA EF7B AFA6
Wake me up when it does something they didn't program it to do.
Just you wait, I have a feeling that half way through the broadcast, Watson will bludgeon the two human opponents to death and declare himself victor. The IBM programmers will deny ever having programmed Watson to do this.
However, Jeopardy answers can be tricky for a machine to decipher, as understanding what exactly is being sought can be difficult linguistically. Often the clues have puns, word-play and other things we humans get easily... NOVA had a really good show about this BTW.
It must have been something you assimilated. . . .
They used an example quote "Last night I shot an elephant in my pajamas"
How did the elephant get in your pajamas?
I don't know if that would be so difficult. You can write an algorithm to diagram a sentence. The problem, in your example, is that there are multiple potential diagrams. (is it "last night, in my pajamas, I shot an elephant" or "last night I shot an elephant in my pajamas"???) Parenthetical elements, and other parts of speech, can be moved around to different parts of a sentence. I would think your algorithm would simply have to compute all possible sentence diagrams and then statistically determine which one is more likely. (for instance, have it use a database of sentence frequency to see which is most common) Then again, I don't really know enough about how they're doing this to know if this makes any sense.
I'm sure IBM had people working on this who have a very extensive language background, though.
I could give you access to any search engine you want, 100 times the computing power and 100 times the amount of time that IBM is using, and you couldn't write a program that could get even 1% of jeopardy questions correct.
This isn't a human assisted program. It is completely automated. And it's fucking amazing.
If Watson's algorithms don't count as intelligence, then human brains don't count as intelligence either. A human brain is a machine. Being human doesn't make you unique or put you above the restrictions of natural law. This tends to upset most humans who compensate by choosing to believe that they have soul, or anything that lets them believe that they are more than just another gear in the machine that is the universe.
This computer is designed to parse non-literal (whatever the term is) phrases filled with idioms and form and answer (or question for Jeopardy). The computer already has all the answers, but needs to figure out what the host is asking for. But, humans and fans work towards answer from the opposite direction. As fans know, ofetn you don't know the answer, but the puns, phraseology, and category often give you hints towards the answer. Not to slight the accomplishments of the programmers, but it doesn't think along the same lines of a real contestant.
I swear to God...I swear to God! That is NOT how you treat your human!
If you'd been on Jeopardy! and told Alex his question was meaningless because John of Gaunt wasn't named John of Gaunt when he was born, I suspect you'd be the one accused of being the computer. Computers give responses that appear over-generalized all the time -- wether the answer is useful depends on the situation, which is why I sorta carefully said "particularly in the domain of how you know trivia." People often over-generalize or give overly-specific answers, too; they often avoid some of the dumber mistakes Watson does, because they have common sense gleaned from decades of high-bandwidth stimulus, but when it comes to abstract rules and concepts, people often are careless and use lessons learned in situations which aren't germane or appropriate to the situation. Look at how little children draw animals, with human eyes and sexual dimorphism. They take what they know and apply it to new things, and produce inaccurate output. These sorts of inconsistencies contain their own internal truth, like "animals are like people" and "the boy turtle thinks the girl turtle is yucky!" and that's the interesting content, because we cannot understand the mind and these outputs are the only window we have into the mind of others. In a similar way, Watson's associations sometimes give you fascinatingly wrong answers, but they reflect an internal truth, like "People in the news talk way too much about 9/11 and anyone that reads enough news will associate it with disparate concepts." This internal truth isn't interesting to anyone, though, because, unlike children, we can see how Watson works.
I mean generally, when someone says "we know the computer doesn't understand something" it's because we understand the computer. Any creature who's cognitive faculty can be disassembled down to its bolts, by definition, cannot understand concepts itself. The only reason we claim that any particular human being "understands" something is because they produce responses that align with our own model of the concept, and because we cannot see how their brain works. I think if you could actually see how people thought about things, with a Thought X-Ray, you'd be shocked at how differently people modeled concepts, to the point where someone with such an ability would conclude that no one really knows anything, and that all knowledge is emergent from collaboration.
Of course I can't prove that, it's all rather fanciful, but that's just as good as asserting a negative, like:
Don't blame me, I voted for Baltar.
No. Your genes are a programming code, and our best information indicates that is EXACTLY what is written in them. Reality could be different...but our best understanding of it today indicates that you are a digital computer with some very sophisticated procedural code.
I have a feeling this isn't going to give us the ultimate question to the ultimate answer...
It's striking how many people are willing to die for things they don't understand, let alone converse about them.
I was struck in Wired for War by the stories of EOD units in Iraq who would name their bomb-defusing robots, give them ranks, promotions, and ribbons, and, touchingly, would mourn their robot's destruction. There's one story about an operator who was literally bawling to a support rep at iRobot, asking if they could please somehow repair their bot. They were real creatures to them, and they were completely unintelligent. What really made the robots alive to them is that they were balky, seemed to have a personality in difficult situations (operator's confirmation bias at work), and had saved the operator's lives many, many times. It didn't matter that the robot didn't "understand" why it was being destroyed, the operators were often in a similar situation... what mattered was its (nominal) selflessness and heroism, something the operator's were required to display as well in a war situation.
I mean like, the Chinese Room is interesting, but the dark secret is that, when it comes to the way human beings confer personhood on other things, it makes it so there is no door to the Chinese Room. Only a mail slot, and it's impossible to see what's on the other side. An an unknowable truth is no truth at all.
Don't blame me, I voted for Baltar.
I know that they can't ring in before Alex has finished reading. My point is that I see it as being problematic that Watson receives the text of the clue electronically at the exact same time that the clue is visually presented to the players. It will take a human reader a few tenths of a second (maybe even as much as a second or two) to actually read the clue, whereas Watson has full knowledge of the text of the clue in mere milliseconds, giving it those few tenths of a second advantage to actually contemplate the clue and possibly arrive at an answer before the humans have even finished reading it. I thus expect that Watson fares poorer on clues which take very little time to read, because it does not have as much of a head start on the human players as it does on the more verbose clues.
File under 'M' for 'Manic ranting'
http://communities.washingtontimes.com/neighborhood/dispatches-heartland/2011/feb/8/ibm-watson-jeopardy-harbinger-vexing-politics/
"It’s hard to imagine the mood of the country changing if these trends persist. Will our bad mood turn to deep funk if we’re entering a period of long-term, systemic underemployment without even a hint of a plan for how we’ll manage? Does it make sense simply to hold on to our faith that eventually life will be like it was in the 1940s, 50s and 60s? It is marvelous to witness the technological advances brought about by human ingenuity. It’s a kick to watch computers compete against humans on shows like Jeopardy. But, IBM Watson’s real success on Jeopardy would be if it helps kick-start a meaningful political conversation about the economy of the future."
Solutions (my comments):
http://econfuture.wordpress.com/2010/10/19/robots-jobs-and-our-assumptions/#comment-392
http://www.journal-of-nuclear-physics.com/?p=360&cpage=6#comment-20270
http://knol.google.com/k/paul-d-fernhout/beyond-a-jobless-recovery/38e2u3s23jer/2
A 21st century issue: the irony of technologies of abundance in the hands of those still thinking in terms of scarcity.
A dying species? We would be kings- with high intelligent slaves to all our work.. this time ethically.
I put that sentence (with the quote marks) into Google, and all it returned was a link to this Slashdot thread.
Damn you Google!
I don't care if it's 90,000 hectares. That lake was not my doing.
Well, scientists definitely do a lot of higher level testing of that sort of thing, with MRIs.
Also, wasn't there another Nova episode recently where they showed an animal learning a specific thing? That is, they had figured out how exactly it learned one new thing.
No, but it will have something substantially better; access to a cloud-based "Watson" that is constantly learning from human interaction just like the current Watson.
Of course, unlike the current Watson, the cloud-based "Watson" will be constantly learning how better to track you down via your cell phone signal, in order to terminate you.
I don't care if it's 90,000 hectares. That lake was not my doing.
I don't think that inputs and outputs are always "sane". That's why things like Escher drawings and optical illusions are so freaky -- they're giving our senses inputs that don't match the physical world that our senses evolved to be able to read.
Also, while we don't halt as in a computer, we do do things like pause momentarily when given confusing input -- e.g. the old footage showing when babies learn about object permanence -- when the moving car/train doesn't come out the other side of the tunnel (the researcher stopped it), the baby gives what I interpret as a "huh?" response.
This is probably because 4 species of grasshopper are the only Kosher insects. Watson obviously knew that, which I didn't! (But I'm not jewish.)
I just watched part 1 of IBM's Watson computer play against two of Jeopardy's best players. Watson finished up round 1 tied with one player at $5K each. Some of the questions that Watson answered were extremely difficult. Watson was able to parse and comprehend answers that were not only lengthy but complicated as well. Despite a couple of Watson's misreads it's performance was extraordinary and IBM researchers deserve a lot of credit for their efforts.
"particularly in the domain of how you know trivia."
I caught your subtlety, but I failed to convey mine. I'm not merely saying that I think differently than a computer because I can generalize concepts. I'm saying that I think differently because I do (constantly) generalize concepts. Sure, I might come up with the trivially correct response to a challenge, but my mind doesn't stop there. I don't know about you, but I'm not a very effective fact machine. I'm not very good at pulling out the correct response to some challenge without dredging up a lot of other related information. My mind will continue to consider the nuances of the issue, explore the conceptual space, possibly to the point of distraction... so much so that I might miss the next challenge, or at least be slow on the buzzer...
The machine, on the other hand would recognize that the required response had been given and that the play was over and it would be ready with a blank slate, waiting for the next challenge.
----------
They take what they know and apply it to new things, and produce inaccurate output. These sorts of inconsistencies contain their own internal truth, like "animals are like people"...
Or "that a computer understands something"...
This is what scares people: maybe intelligence isn't so hard after all.
This is my opinion: a great deal of what passes for "intelligence" in the world is just language processing and memory. Not so much "figuring things out" as much as knowing where to look for answers and applying the result. Not creative thinking, but the use of what we already know, or, with language, what some other bloke already knows.
I see a lot of comments trying to downplay this event but, look: if Watson could beat you at this game, how smart are you, really?
Honesty. Loyalty. Kindness. Laughter. Generosity. Magic!
"What would be a good definition artificial intelligence that wouldn't be subject to goalpost moving?"
When the AI starts chiming in on where the goalpoast is, and giving us suggestions, I think then we can hang our hats and go home as a dying species.
You aren't intelligent because you can think about what being intelligent means, you are intelligent because you can think about what Douglas Hofstadter thinks intelligence means.
That seems to presuppose things about consciousness and intent that I see no reason to believe would be any different for humans than for machines. In particular, I don't see any reason not to describe a human as a biological machine.
Maybe if we started by sufficiently defining "understanding," rather than relying on intuition, it'd be a stronger claim.
Don't thank God, thank a doctor!
You're making a *huge* assumption that humans operate algorithmically, which is exactly the problem being discussed in the debate between Searle, Dennett, Haugeland, and others. Getting a handle on "understanding", or more specifically the origin of meaning, is the larger question that these folks are addressing. So you're entirely correct about needing more than intuition, and that is what people have been trying to do using these means for many, many years.
4096R/EF7BAFA6 79E1 DF98 D09D 898F 9A11 F6F0 DDDC 23FA EF7B AFA6
Yes, a very good point. AI doesn't need to be smarter than All People, only most people. And I think the brute force simulations of it are getting pretty close. Between AI simulations and robots (and all the acoutrements like sensor tech and such) we're rapidly being obsoleted. The question then is what next? Space?
Cool! Amazing Toys.
Hey, Google has an R&D! And at Google you get to eat free food! And you get props if you make something cool! Maybe Sergey will let you post your HTML5 Pac Man to the front page! To remind everyone, especially the shareholders, that we're new money! Yeah, that'll pan out long term!
Having worked with IBM in the past, I agree, is what I'm saying. At Big Blue they all still wear suits and button up shirts though. And conduct themselves in a professional way. I think that's the thing about IBM I respect the most, they've been in business for over 100 years doing basically the same thing and they are consummate professionals. Building the engines that power the economies of the world. Google is just riding on their coat tails, economically speaking. If it wasn't for our massive consumption economy there wouldn't be a Google because there wouldn't be ads. But there would still be IBM because someone has to count the taxes or rations or whatever. That being said, I'm happy Google exists, it's just a different sort of game that I don't take that seriously as an intellectual with an understanding of the true impact of technology. And that's a lot more than market cap, flashy screens, consumers. It's the ability to help organize societies, capitalize on specialization, ensure resources are allocated quickly and efficiently. With REAL THINGS, not just web pages. Arguably these could be seen as tools of control, and in some sense they are, but it's essential to society as a whole to have some sort of structure that can guide us. And you don't think Google is controlling you when they decide what a "quality page" is? Or decide what's evil and what isn't for you? Googles, Apples, Microsofts have come and gone but big blue will always be there, no matter what. Oh, and Lehman Brothers. Hypocritical circular logic all in one post!
Cool! Amazing Toys.
Why did it take "motorcycle club" into account in at least three of the "M.C. 5" questions? (reference: http://www.youtube.com/watch?feature=player_detailpage&v=WFR3lOm_xhE#t=125s)
1266953+17
It's a rather large assumption, but it does seem like a reasonable default assumption, given what we know about how humans *do* operate.
I realize it's a somewhat controversial opinion, though I'm not entirely sure why. For one, it seems weak determinism is likely true, or that there's some element of randomness, but computer can be random also. Even if that wasn't the case, it seems like the alternative to anything algorithmic is necessarily a dualist stance, and that seems like a pretty big assumption as well.
Essentially, what I'm doing here is a very liberal application of Occam's Razor. It really does seem like the only options are an algorithmic reality or a dual one.
Don't thank God, thank a doctor!
In fact, the assumption that cognition is computational or mechanistic is where the mind-body problem comes from (right from the 17th century). If it's all just syntactic, then explaining semantics is now a problem, and you have problems like dualism. Non-algorithmic doesn't mean magic, by any stretch. If you don't assume algorithmic, then there is no mind-body separation about which you can have a dualist stance.
If you think that the universe in general is algorithmic, e.g. that the evolution of the sun is an algorithmic process, then we might easily only disagree about terminology. Those who do not follow computational theories of mind might say that the complexities of intelligent behavior are more like the complexities of the sun than anything you will get out of a turing machine.
There are a frightening number of different issues once we go into details, and there is a long history of people attempting to address them. I don't think a slashdot thread is capable of holding it all. And of course, the more one learns the more one realizes that the answers are far fewer than the questions.
4096R/EF7BAFA6 79E1 DF98 D09D 898F 9A11 F6F0 DDDC 23FA EF7B AFA6
Yeah, but unless we're interacting with it remotely, we won't each have Watson's capability in our hands soon.. (Cue Moore's Law references...)
fear of a robot planet.
My cousin is a pretty smart guy (he has numerous patents in optelectronics ) and worked at HP Labs before they spun out into Agilent.
He said it was a regular occurrence to have his budget ( for optical research) slashed/raided to give more money to the Printer division
who needed new plastic molds for the next model of Inkjet.
That was the main reason he left, even if it was the correct investment from a shareholder perspective, it was insulting to him to
see the low priority placed on research.
freakonomics is reporting that watson won on day 3.
February 18, 2011, 9:30 am
It’s Official: The Computer’s Smarter
By FREAKONOMICS
The IBM supercomputer named Watson has beaten two Jeopardy! champions in a three-night marathon. The computer was awarded a $1 million prize, but the BBC reports that “the victory for Watson and IBM was about more than money. It was about ushering in a new era in computing where machines will increasingly be able to learn and understand what humans are really asking them for. Jeopardy is seen as a significant challenge for Watson because of the show’s rapid-fire format and clues that rely on subtle meanings, puns, and riddles; something humans excel at and computers do not.” With his final answer, Ken Jennings, one of the human competitors and the winner of 74 consecutive Jeopardy! shows (a record), wrote, “I for one welcome our new computer overlords.”
"Am I self-aware?"
;p
Oh, wait, answering questions does not demonstrate much...
One that hath name thou can not otter