Think low-tech. Remember the horror in Les Nessman's voice as he broadcast the results of WKRP's Thanksgiving Day promotion? The sight of turkeys dropped from the sky was at first festive, but grew terrifying as the WKRP staff realized that turkeys can't fly. The resulting thuds and splats would scare enemies just as they scared the innocent residents of Cincinnati that dark day.
It's a great choice. You'll kill the guys working in x86 assembler.
Re:Mercury (Perl has first-class functions)
on
ICFP 2001 Task
·
· Score: 2
No, Perl has too many capabilities to be considered a purely functional language, however it does have first-class functions, so any program that can be created in more than one functional language (say Lisp and ML) can also be created in Perl
Lisp is not just a functional language, as any functional purist will tell you with an air of disdain. It supports multiple paradigms.
I don't know Perl, so the following is a real question, not merely rhetorical: Can you create new functions on the fly in Perl as you can in Lisp? That is, can you have a function1 that returns a new function2 as its value, where function2's definition is relative to the parameters that were passed into function1 at its time of creation? That is, can you create closures in Perl? In Lisp, this is done as follows:
Just passigning functions around does not make a language functional. Heck, C can do this function pointer. Truly functional languages support the creation of closures (among other things). Can Perl do this?
There's theft and then there's theft. Everyone draws their own line. I've seen two movies for the price of one and felt no guilt. I've also walked out of shit movies and received no reimbursement. And, on occasion, I've had to leave movies because of loud patrons and been compensated with a single free pass (come back another day and try again!) but not the time I wasted on the aborted effort.
Hank Azaria [...] stole the show and is worth seeing because of him.
Agreed. He was brilliant. Didn't he also do the memorable 'Agador Spartacus' in 'The Birdcage'? That was another great performance.
For eye candy, you can peep Catherine-Zeta Jones and Julia Roberts.
Catherine Zeta-Jones is indeed stunning.
I also thought she maximized her character. She was dead-on as a Hollywood prima donna bitch. But Zeta-Jones, Azaria, and Alan Arkin, and Walken had the only well-written parts in the movie. And of this bunch, only Zeta-Jones had significant screen time. The rest of the actors -- Cusack, Roberts, and Crystal -- walked around the same goddam hotel for two hours spewing what were, to my ears, non-sequitir after non-sequitir. Plus, the series of 'revelations' after the screening are beyond belief, as is Cusack's attraction to the utterly boring Roberts. Billy Crystal's script and direction utterly fail.
By the way, Roberts fans (not me) will be disappointed by this movie. She dresses poorly, shows no radiance, and her glorious smile is nearly absent. Of course, this is in line with her dowdy character, but still. Julia Roberts without the girly-girl-isms is just another anorexic brown-haired woman.
Me, I had a bad experience at a burger king in college. I haven't set foot in a burger king in eight years now.
And I thought I was the only one. My last year in college, I ordered a chicken sandwich. I bit into it and noticed a problem -- it was too crispy and light. Pulled it apart and discovered the oddest thing. The breading for the chicken was there, but inside there was no chicken!. They replaced it with a normal one but no explanation was offered for how a breading 'sleeve' could wind up on my sandwich wrapping nothing.
I haven't eaten a chicken sandwich since then -- about ten years ago.
The thing that gets me is this: I always assumed that breaded patties were made the old fashioned way -- someone grabbing a piece of meat, dragging it through some breadcrumbs, and tossing it in fryer. But such a process cannot break down in a way that yields an empty sleeve of breading. My nightmares consisted, for a while, of a factory with a Y-shaped assembly line. Down one line came soap-bar-shaped slabs of processed meat, down the other empty sleeves of breading. At the intersection was a fat sweating man. His job was to mate meat with breading, stuffing the former into the latter and sealing the result somehow. He must miss now and then...
Did anyone else think that the Google translation of the German original was poetically mystifying,? It sounded like the pronouncements of a shy, conquering alien. Consider:
No costs are to arise for the KIllustrator developer
One wants to achieve an amicable agreement now.
Such a regulation - change of name against retreat of the warning
Adobe regrets now that [...] annoyance resulted to the KIllustator developer.
The primary problem with the net2phone box is that it uses the Internet as transport, and there is no end-to-end management of the internet. So, the packets that contain your voice have to compete on an equal footing with other peoples packets, despite the fact that yours are much more time-sensitive.
Check out this article for information of smart routers that prioritize packets based on their contents. The claim is that packets containing temporal information (e.g., audio, video) will be passed more synchronously that packets containing less temporally-dependent information.
Yeah, I agree! Yay for China standing up for itself.
Retard, did you read the second paragraph of my post? It included this caveat:
Please don't interpret this comment as condoning the actions of the Chinese government against its own dissidents and the peoples of other countries, its selling of arms to 'rogue' states, etc. That's not my intent.
I complimented them for showing some balls against Western dictates; I didn't condone their human rights violations.
By the way, I followed your interesting link. Pretty appalling stuff. Once again, NOT TO CONDONE CHINA'S ACTIONS, but did you notice that all of the skinning was of executed prisoners -- i.e., corpses -- for transplant to burn victims except for one (admittedly gruesome) case where the executed prisoner had not finished dying?
I kind of admire China on this one. It's a big, powerful, Asian country that does things its own way, and this just infuriates the West to no end. You can make your little laws and jump up and down as much as you want but at the end of the day, we're gonna do what we want. And you know what? Whether their actions are utlimately right or wrong, at the end of the day, the West always capitulates, always loses face. Why? Because we're greedy bastards who don't want to lose access to their billion customers.
Please don't interpret this comment as condoning the actions of the Chinese government against its own dissidents and the peoples of other countries, its selling of arms to 'rogue' states, etc. That's not my intent. I simply wish to point out that China does not let the West bully it; if anything, it bullies the West. None of the other Asian countries can make the same claim. (Note: I'm of Indian descent.)
Personally, I like people that make up words as it shows a sign of creativity. I'd rather have someone in office who mixes up words sometimes and is somewhat creative, than yet another bland politician type.
Only if you are the kind of person that thinks that any attack on racism is in itself, racist.
No sane person thinks this.
In responding to an ignorant post about Iraquis (in general), an equally ignorant post about Americans (in general) was made. That is, an ignorant, stereoptypical comment about a class of people was countered with an ignorant, stereotypcial comment about another class of people. Do you see the irony? I would have remained silent if the poster had said 'you (specifically) are a moron (if you characterize all Iraquis this way)' rather than 'you and all your kind are morons'.
Do you see the difference?
Hmmm. Then again, there is a certain truth to your logic: 'Any attack on a Slashdot post is itself a Slahdot post.'
I want to see this guy working on The Revolution. Is there a bio for him? Do they have a "current life status" [...]? Do they have him hooked up to record vital statistics?
His name is Nasubi.
He seems like a likable if odd fellow.
Minsky was just arguing that a perceptron could not compute an XOR function, and the reason he was wrong is simply because he didn't consider that you might connnect one perceptron to the output of another.
For Minsky to not even consider connected perceptrons was a humungous brain fart for which he should rightly be ridiculed, particularly given the influential position the way he was in at the time, and the effect it had stifling all ANN funding and research for a long time.
I think you have your facts wrong. Minsky and Papert proved that single perceptrons could only represent linearly separable mappings, and thus couldn't represent XOR for example. They also proved (or perhaps it was already known at the time since it's kind of obvious) that networks of units with linear activation functions collapse to single perceptrons, and thus suffered the same representational limitation.
They went on to express skepticism that an algorithm for training networks of units with nonlinear units to represent linearly inseparable mappings (such as the XOR function) would be discovered. On this conjecture, they were wrong.
That is, training algorithms such as backpropogation allow networks with at least one layer of sufficiently many hidden units, where each unit has a nonlinear activation function, to usually represent arbitrary mappings. I say 'usually' because gradient descent techniques like back propogation are not guaranteed immunity from local minima, although in practice this does not appear to be a problem.
There are other training algorithms for neural networks for which I am less knowledgable -- they may not suffer even in principle from local minima. There are also proofs (like Judd's) about the intractability of optimally 'loading' nontrivial neural networks. In other words, I don't know the current state of the art.
But you are clearly wrong in what you think Minsky proved and what he conjectured. That his conjecture, built on his own research, has proven largely incorrect is just a testament to the unpredictability of science. Connectionists need to stop with their conspiracy theory malarky. Minsky's no bogeyman either.
The truth is that the brain is a collection of neural nets feeding into and relying on one another.
Neural nets are for sissies. If you're gonna reduce cognition, why stop at the level of neural works? Everyone knows that biology and chemistry are just convenient ways to talk about particle physics. Real scientists are writing the equations of AI in terms of quarks.
You probably use high-level programming languages like C and assembler instead of directly writing binary machine language. Sheesh!
Yeah, but the symbolic part of it itself isn't a hard problem - Allen Newell's SOAR already does pretty much everything you could hope.
As well, John Anderson's ACT-R (http://act.psy.cmu.edu/) system. Both Soar and ACT-R have been implemented as connectionist networks to illustrate that, for the Church-Turing challenged, one can construe cognition as symbolic and simultaneously assume a biological implementation in the brain. (That is, if you think connectionist networks are neurally plausible, which many -- such as Noble laureates Francis Crick and Gerald Edelman, I believe -- do not.)
It is interesting to note that both ACT-R and Soar can be augmented with perceptual and motor subsystems, built consistently with the psychological data on perception and action, to form a truly unified cognitive architecture. The augmented systems have been used to simulate real-world cognition such as flying Air Force fighters (see Rosenbloom's work at USC), playing Quake (see Laird's work at Michigan), and simulating an air traffic controller (see Anderson's work at Carnegie Mellon).
Any neural network models out here capable of such authentic cognition, please step forward. (Except Dean Pomerleau's Autonomous Land Vehicle -- you're the exception that proves the rule!)
The mention of Soar is also a good lead in to debunking a persistent myth about connectionist networks -- that they are the only systems that learn. There are a plethora of learning algorithms for symbolic architectures too, as well as statistical algorithms, genetic algorithms, etc. Don't believe the connectionist hype! For example, Doorenbos showed in the mid 1990s that Soar models could learn a million rules with no slow down in performance; these million-rule models ran (and continued to learn) efficiently on Sun workstations of the time. His CMU dissertation is a masterpiece, for those interested.
The Minksy "Agent" model is very much on the "Connectionist" side of the map.
To echo this important point, Minsky's "Society of Mind" is as much like an "interactive" connectionist network as it is a symbolic architecture. For example, Hofstadter cites the SOM approach and Holland's genetic algorithms (as well as the highly interactive HEARSAY system) as precursors of and inspirations for his own work, which is by no means symbolic.
Marvin Minsky is one of the reasons that AI still has not delivered on its promises.
Are you trolling or just delusional? AI has not delivered in its promises in part because it's a damned hard problem. Philosophers, psychologists, linguists, and others have pursued it for 100s of years, and for most of its 50 year existence in computer science departments, it has been pursued on pathetically underpowered machines. The field's in its embryonic stage. Hell, ANNs have come and gone twice already (i.e., with McCulloch/Pitts/Hebb and then Rosenblatt); will you sour on ANN-AI if the third time is not a charm? I hope not, oh faithless one.
He is part of the old symbolic school of AI.
Is the 'old' or the 'symbolic' part your rhetorical way of discrediting him? First of all, as I pointed out in an earlier comment, it is plain wrong to put Minsky in the same bin as McCarthy, Chomsky, and other 'neats' who believe that logic is the royal road to AI. Minsky is a scruffy, an "anything goes" type. Second, do you know what he Minsky did for his dissertation circa 1950 or so? Built the first physical ANN out of primitive hardware in the basement of the math building at Princeton. That's right, he's an old skool ANN researcher. He also came by his skepticism the old fashioned way -- he eeeaaarrrned it. He doesn't simply oppose ANNs for philosophical reasons, brain boy.
He was the guy who, with Seymour Papert, wrote a scathing criticism of the then embryonic field of neural networks, effectively strangling research in neural networks for the better part of a decade.
He cowrote a book that formally disproved claims that irresponsible ANN researchers of the time were making about the representational capacities of their nets and inductive powers of their learning algorithms. He did express skepticism that these could be overcome. He was proven wrong on this count 15 years later (not counting Werbos's work). Again, you make it seem like he was an irrational crusader against ANNs. He was not. From experience and mathematical analysis, he made a conjecture. He was wrong. Such is the scientific life.
Frankly, I've never understood why ANN types cry and complain about the lack of ANN funding in the 1970s. Given analyses like those of Minsky and Papert, and the successes of the symbolic approaches of the time (ever heard of HEARSAY and how Rumelhart and Hofstadter claim it as a primary inspiration?), it appeared that ANN research would not be as fruitful as symbolic approaches. Hindsight is 20-20 on this one of course -- ANNs are better than was thought then. But hindsight always works this way. Why exactly did ANN researchers give up? "Oh boo hoo, not as much money for my ANN work -- must follow other fads." Their lack of resiliency was cowardly. (I feel the same way about symbolic folks who lie shamed in the shadows right now.)
IOW, what we really need to understand is the learning process, which encompasses perceptual, motivational and motor learning.
This is the main reason why we still don't have human-level AI! I think Minsky's stance is a disservice to computational neuroscience and ANN researchers everywhere.
We need to understand learning and development, perceptual and motor activity (not just learning, but performance too), how all of this is coupled with cognition (unless you think, for example, Cantor's Continuum Hypothesis is best represented perceptual-motor-ly), etc. We also need to understand how all of this is embodied in the body more generally, the ecological environment, and the culture at large. ANNs show little if any promise at tackling this last set of questions. (Journal article we'll never see: "An ANN analysis of female circumcision"). So get off your high horse about how ANN folks are gonna crack the whole cognitive nut. Frankly, it's not at all clear that ANNs are the right formalism for investigating even learning and development. A number of developmental neurosciences favor other approaches (e.g., dynamical systems).
The man has had his day in the sun. Now it's time for the younger generation of AI researchers to come in and say "hold it! we're taking a different approach from now on. The unkept promises of AI were made by the old symbolic AI crowd. There is a new school in town. The new AI neural, it's emergent, and it's gonna to kick ass!"
Oh, you're just the AI equivalent of a script kiddie. Why did I bother?
The idea that man is equal to God, and believes he can put himself on an equal footing with the Divine Creator himself is just the sort of ridiculous notion that could only come from the USA.
Funny stuff.
I also find it humorous that men, in equating themselves with the Divine creator himself, have even attempted to decode his secrets through their studies of such dubious subjects as physics, chemistry, and biology. It is no wonder these areas have scarcely progressed in their millenia of existence. Compared to physical scientists, the audacity of AI researchers like Minsky has been brief in its duration.:-)
For those who would equate AI with symbolic approaches and Lisp, and the widely-believed failure of symbolic AI with the puported inadequacy of Lisp, note that Brooks is a Lisp guy and at least his original subsumption work relied on the Scheme programming language (a variant of Lisp), perhaps implemented directly in a chip.
(Note that the poster I'm replying to didn't make these inferences -- this is just a convenient place for me to attach this info.)
I have an impression that according to Minsky mind is something completely authonomous, with its own absolute (logical rules), something you can implement in a machine (hence the whole notion of AI).
Actually, Minsky doesn't believe in 'logical rules'. The Stanford group (following McCarthy) believes that logic is the royal road to AI. For this, they are termed the 'neats' and contrasted with the more pragmatic 'scruffies' such as the folks under Minsky's direction at MIT. Incidentally, there are 'neats' at MIT too -- for example, Chomsky -- who are most definitely at odds with Minsky and his, well, Minions.
There is another point of view (which I believe is more adequate) which boils down to mind and intellect as fairly sophisticated adaptation function (or tool), so to implement AI we have to start with very simple machine capable of interacting with its environment, learning, adapting, and evolving.
What makes you think this is not AI? Logic, Minsky's work, Brooks's robots, genetic algorithms -- these are all approaches to AI. The domain of AI is not isomorphic with logic.
Minsky separates the mind from the body, from the environment, from the whole human exprience, and this is plainly wrong.
Well this is your opinion, as well as that of the 'embodied cognition' folks and other latter-day phenomenologists. It may or may not be correct. Frankly, the embodied approach has not produced the great AI breakthrough that has eluded classical approaches.
Perhaps all of these paradigmatic battles and half-truths are the poisonous work of philosophers? Nah, couldn't be.
Linux is good for those who value their freedom (speech and beer)
Your comment is tongue and cheek, I know, but just to be clear: Linux is for those who value the freedom to see the source -- that's it. There are other kinds of freedom, and I doubt you can make platform-dependent generalizations for these.
Back to the article: Of course such comparisons are meaningless, but I found it interesting that they gave the Interface comparison to Windows (by a nose). Their argument was essentially 'we know the Windows interface, it remains unchanged, therefore it's better'. By this logic, OS X will only win when it becomes more familiar (read: Windows-like). Weird.
Slashdot Mirror
Think low-tech. Remember the horror in Les Nessman's voice as he broadcast the results of WKRP's Thanksgiving Day promotion? The sight of turkeys dropped from the sky was at first festive, but grew terrifying as the WKRP staff realized that turkeys can't fly. The resulting thuds and splats would scare enemies just as they scared the innocent residents of Cincinnati that dark day.
why would C be inferior for solving this task?
It's a great choice. You'll kill the guys working in x86 assembler.
Lisp is not just a functional language, as any functional purist will tell you with an air of disdain. It supports multiple paradigms.
I don't know Perl, so the following is a real question, not merely rhetorical: Can you create new functions on the fly in Perl as you can in Lisp? That is, can you have a function1 that returns a new function2 as its value, where function2's definition is relative to the parameters that were passed into function1 at its time of creation? That is, can you create closures in Perl? In Lisp, this is done as follows:
Just passigning functions around does not make a language functional. Heck, C can do this function pointer. Truly functional languages support the creation of closures (among other things). Can Perl do this?
There's theft and then there's theft. Everyone draws their own line. I've seen two movies for the price of one and felt no guilt. I've also walked out of shit movies and received no reimbursement. And, on occasion, I've had to leave movies because of loud patrons and been compensated with a single free pass (come back another day and try again!) but not the time I wasted on the aborted effort.
Hank Azaria [...] stole the show and is worth seeing because of him.
Agreed. He was brilliant. Didn't he also do the memorable 'Agador Spartacus' in 'The Birdcage'? That was another great performance.
For eye candy, you can peep Catherine-Zeta Jones and Julia Roberts.
Catherine Zeta-Jones is indeed stunning.
I also thought she maximized her character. She was dead-on as a Hollywood prima donna bitch. But Zeta-Jones, Azaria, and Alan Arkin, and Walken had the only well-written parts in the movie. And of this bunch, only Zeta-Jones had significant screen time. The rest of the actors -- Cusack, Roberts, and Crystal -- walked around the same goddam hotel for two hours spewing what were, to my ears, non-sequitir after non-sequitir. Plus, the series of 'revelations' after the screening are beyond belief, as is Cusack's attraction to the utterly boring Roberts. Billy Crystal's script and direction utterly fail.
By the way, Roberts fans (not me) will be disappointed by this movie. She dresses poorly, shows no radiance, and her glorious smile is nearly absent. Of course, this is in line with her dowdy character, but still. Julia Roberts without the girly-girl-isms is just another anorexic brown-haired woman.
Me, I had a bad experience at a burger king in college. I haven't set foot in a burger king in eight years now.
And I thought I was the only one. My last year in college, I ordered a chicken sandwich. I bit into it and noticed a problem -- it was too crispy and light. Pulled it apart and discovered the oddest thing. The breading for the chicken was there, but inside there was no chicken!. They replaced it with a normal one but no explanation was offered for how a breading 'sleeve' could wind up on my sandwich wrapping nothing.
I haven't eaten a chicken sandwich since then -- about ten years ago.
The thing that gets me is this: I always assumed that breaded patties were made the old fashioned way -- someone grabbing a piece of meat, dragging it through some breadcrumbs, and tossing it in fryer. But such a process cannot break down in a way that yields an empty sleeve of breading. My nightmares consisted, for a while, of a factory with a Y-shaped assembly line. Down one line came soap-bar-shaped slabs of processed meat, down the other empty sleeves of breading. At the intersection was a fat sweating man. His job was to mate meat with breading, stuffing the former into the latter and sealing the result somehow. He must miss now and then...
Did anyone else think that the Google translation of the German original was poetically mystifying,? It sounded like the pronouncements of a shy, conquering alien. Consider:
No costs are to arise for the KIllustrator developer
One wants to achieve an amicable agreement now.
Such a regulation - change of name against retreat of the warning
Adobe regrets now that [...] annoyance resulted to the KIllustator developer.
The primary problem with the net2phone box is that it uses the Internet as transport, and there is no end-to-end management of the internet. So, the packets that contain your voice have to compete on an equal footing with other peoples packets, despite the fact that yours are much more time-sensitive.
Check out this article for information of smart routers that prioritize packets based on their contents. The claim is that packets containing temporal information (e.g., audio, video) will be passed more synchronously that packets containing less temporally-dependent information.
Yeah, I agree! Yay for China standing up for itself.
Retard, did you read the second paragraph of my post? It included this caveat:
Please don't interpret this comment as condoning the actions of the Chinese government against its own dissidents and the peoples of other countries, its selling of arms to 'rogue' states, etc. That's not my intent.
I complimented them for showing some balls against Western dictates; I didn't condone their human rights violations.
By the way, I followed your interesting link. Pretty appalling stuff. Once again, NOT TO CONDONE CHINA'S ACTIONS, but did you notice that all of the skinning was of executed prisoners -- i.e., corpses -- for transplant to burn victims except for one (admittedly gruesome) case where the executed prisoner had not finished dying?
I kind of admire China on this one. It's a big, powerful, Asian country that does things its own way, and this just infuriates the West to no end. You can make your little laws and jump up and down as much as you want but at the end of the day, we're gonna do what we want. And you know what? Whether their actions are utlimately right or wrong, at the end of the day, the West always capitulates, always loses face. Why? Because we're greedy bastards who don't want to lose access to their billion customers.
Please don't interpret this comment as condoning the actions of the Chinese government against its own dissidents and the peoples of other countries, its selling of arms to 'rogue' states, etc. That's not my intent. I simply wish to point out that China does not let the West bully it; if anything, it bullies the West. None of the other Asian countries can make the same claim. (Note: I'm of Indian descent.)
Personally, I like people that make up words as it shows a sign of creativity. I'd rather have someone in office who mixes up words sometimes and is somewhat creative, than yet another bland politician type.
George W. Bush, the world's oldest script kiddie.
hmmm... that was kinda the point.
D'oh! I should have taken your login name (SubtleNuance) more literally.
Only if you are the kind of person that thinks that any attack on racism is in itself, racist.
No sane person thinks this.
In responding to an ignorant post about Iraquis (in general), an equally ignorant post about Americans (in general) was made. That is, an ignorant, stereoptypical comment about a class of people was countered with an ignorant, stereotypcial comment about another class of people. Do you see the irony? I would have remained silent if the poster had said 'you (specifically) are a moron (if you characterize all Iraquis this way)' rather than 'you and all your kind are morons'.
Do you see the difference?
Hmmm. Then again, there is a certain truth to your logic: 'Any attack on a Slashdot post is itself a Slahdot post.'
I want to see this guy working on The Revolution. Is there a bio for him? Do they have a "current life status" [...]? Do they have him hooked up to record vital statistics?
His name is Nasubi. He seems like a likable if odd fellow.
If Joe-Ignorant-Myopic-Rednecked-Yankee can; why Cant our friend Abdul?
Ohhhh - i forgot, American propaganda doesnt tell Abdul he lives in the height of humanity and should look down on others with disdain.
You're just as ignorant and guilty of stereotypical thinking (against 'Americans', not Iraquis) as the original poster.
Minsky was just arguing that a perceptron could not compute an XOR function, and the reason he was wrong is simply because he didn't consider that you might connnect one perceptron to the output of another.
For Minsky to not even consider connected perceptrons was a humungous brain fart for which he should rightly be ridiculed, particularly given the influential position the way he was in at the time, and the effect it had stifling all ANN funding and research for a long time.
I think you have your facts wrong. Minsky and Papert proved that single perceptrons could only represent linearly separable mappings, and thus couldn't represent XOR for example. They also proved (or perhaps it was already known at the time since it's kind of obvious) that networks of units with linear activation functions collapse to single perceptrons, and thus suffered the same representational limitation.
They went on to express skepticism that an algorithm for training networks of units with nonlinear units to represent linearly inseparable mappings (such as the XOR function) would be discovered. On this conjecture, they were wrong.
That is, training algorithms such as backpropogation allow networks with at least one layer of sufficiently many hidden units, where each unit has a nonlinear activation function, to usually represent arbitrary mappings. I say 'usually' because gradient descent techniques like back propogation are not guaranteed immunity from local minima, although in practice this does not appear to be a problem.
There are other training algorithms for neural networks for which I am less knowledgable -- they may not suffer even in principle from local minima. There are also proofs (like Judd's) about the intractability of optimally 'loading' nontrivial neural networks. In other words, I don't know the current state of the art.
But you are clearly wrong in what you think Minsky proved and what he conjectured. That his conjecture, built on his own research, has proven largely incorrect is just a testament to the unpredictability of science. Connectionists need to stop with their conspiracy theory malarky. Minsky's no bogeyman either.
The truth is that the brain is a collection of neural nets feeding into and relying on one another.
Neural nets are for sissies. If you're gonna reduce cognition, why stop at the level of neural works? Everyone knows that biology and chemistry are just convenient ways to talk about particle physics. Real scientists are writing the equations of AI in terms of quarks.
You probably use high-level programming languages like C and assembler instead of directly writing binary machine language. Sheesh!
Your post is fluff, but I found this line particularly errorful, and thus amusing:
The day is dawning when more people will realize that to "understand" does not always mean to have a linear, provable trail of induction.
Come on over to the logical side, where they use the incomplete tool of deduction!
Yeah, but the symbolic part of it itself isn't a hard problem - Allen Newell's SOAR already does pretty much everything you could hope.
As well, John Anderson's ACT-R (http://act.psy.cmu.edu/) system. Both Soar and ACT-R have been implemented as connectionist networks to illustrate that, for the Church-Turing challenged, one can construe cognition as symbolic and simultaneously assume a biological implementation in the brain. (That is, if you think connectionist networks are neurally plausible, which many -- such as Noble laureates Francis Crick and Gerald Edelman, I believe -- do not.)
It is interesting to note that both ACT-R and Soar can be augmented with perceptual and motor subsystems, built consistently with the psychological data on perception and action, to form a truly unified cognitive architecture. The augmented systems have been used to simulate real-world cognition such as flying Air Force fighters (see Rosenbloom's work at USC), playing Quake (see Laird's work at Michigan), and simulating an air traffic controller (see Anderson's work at Carnegie Mellon).
Any neural network models out here capable of such authentic cognition, please step forward. (Except Dean Pomerleau's Autonomous Land Vehicle -- you're the exception that proves the rule!)
The mention of Soar is also a good lead in to debunking a persistent myth about connectionist networks -- that they are the only systems that learn. There are a plethora of learning algorithms for symbolic architectures too, as well as statistical algorithms, genetic algorithms, etc. Don't believe the connectionist hype! For example, Doorenbos showed in the mid 1990s that Soar models could learn a million rules with no slow down in performance; these million-rule models ran (and continued to learn) efficiently on Sun workstations of the time. His CMU dissertation is a masterpiece, for those interested.
The Minksy "Agent" model is very much on the "Connectionist" side of the map.
To echo this important point, Minsky's "Society of Mind" is as much like an "interactive" connectionist network as it is a symbolic architecture. For example, Hofstadter cites the SOM approach and Holland's genetic algorithms (as well as the highly interactive HEARSAY system) as precursors of and inspirations for his own work, which is by no means symbolic.
Marvin Minsky is one of the reasons that AI still has not delivered on its promises.
Are you trolling or just delusional? AI has not delivered in its promises in part because it's a damned hard problem. Philosophers, psychologists, linguists, and others have pursued it for 100s of years, and for most of its 50 year existence in computer science departments, it has been pursued on pathetically underpowered machines. The field's in its embryonic stage. Hell, ANNs have come and gone twice already (i.e., with McCulloch/Pitts/Hebb and then Rosenblatt); will you sour on ANN-AI if the third time is not a charm? I hope not, oh faithless one.
He is part of the old symbolic school of AI.
Is the 'old' or the 'symbolic' part your rhetorical way of discrediting him? First of all, as I pointed out in an earlier comment, it is plain wrong to put Minsky in the same bin as McCarthy, Chomsky, and other 'neats' who believe that logic is the royal road to AI. Minsky is a scruffy, an "anything goes" type. Second, do you know what he Minsky did for his dissertation circa 1950 or so? Built the first physical ANN out of primitive hardware in the basement of the math building at Princeton. That's right, he's an old skool ANN researcher. He also came by his skepticism the old fashioned way -- he eeeaaarrrned it. He doesn't simply oppose ANNs for philosophical reasons, brain boy.
He was the guy who, with Seymour Papert, wrote a scathing criticism of the then embryonic field of neural networks, effectively strangling research in neural networks for the better part of a decade.
He cowrote a book that formally disproved claims that irresponsible ANN researchers of the time were making about the representational capacities of their nets and inductive powers of their learning algorithms. He did express skepticism that these could be overcome. He was proven wrong on this count 15 years later (not counting Werbos's work). Again, you make it seem like he was an irrational crusader against ANNs. He was not. From experience and mathematical analysis, he made a conjecture. He was wrong. Such is the scientific life.
Frankly, I've never understood why ANN types cry and complain about the lack of ANN funding in the 1970s. Given analyses like those of Minsky and Papert, and the successes of the symbolic approaches of the time (ever heard of HEARSAY and how Rumelhart and Hofstadter claim it as a primary inspiration?), it appeared that ANN research would not be as fruitful as symbolic approaches. Hindsight is 20-20 on this one of course -- ANNs are better than was thought then. But hindsight always works this way. Why exactly did ANN researchers give up? "Oh boo hoo, not as much money for my ANN work -- must follow other fads." Their lack of resiliency was cowardly. (I feel the same way about symbolic folks who lie shamed in the shadows right now.)
IOW, what we really need to understand is the learning process, which encompasses perceptual, motivational and motor learning.
This is the main reason why we still don't have human-level AI! I think Minsky's stance is a disservice to computational neuroscience and ANN researchers everywhere.
We need to understand learning and development, perceptual and motor activity (not just learning, but performance too), how all of this is coupled with cognition (unless you think, for example, Cantor's Continuum Hypothesis is best represented perceptual-motor-ly), etc. We also need to understand how all of this is embodied in the body more generally, the ecological environment, and the culture at large. ANNs show little if any promise at tackling this last set of questions. (Journal article we'll never see: "An ANN analysis of female circumcision"). So get off your high horse about how ANN folks are gonna crack the whole cognitive nut. Frankly, it's not at all clear that ANNs are the right formalism for investigating even learning and development. A number of developmental neurosciences favor other approaches (e.g., dynamical systems).
The man has had his day in the sun. Now it's time for the younger generation of AI researchers to come in and say "hold it! we're taking a different approach from now on. The unkept promises of AI were made by the old symbolic AI crowd. There is a new school in town. The new AI neural, it's emergent, and it's gonna to kick ass!"
Oh, you're just the AI equivalent of a script kiddie. Why did I bother?
The idea that man is equal to God, and believes he can put himself on an equal footing with the Divine Creator himself is just the sort of ridiculous notion that could only come from the USA.
:-)
Funny stuff.
I also find it humorous that men, in equating themselves with the Divine creator himself, have even attempted to decode his secrets through their studies of such dubious subjects as physics, chemistry, and biology. It is no wonder these areas have scarcely progressed in their millenia of existence. Compared to physical scientists, the audacity of AI researchers like Minsky has been brief in its duration.
For those who would equate AI with symbolic approaches and Lisp, and the widely-believed failure of symbolic AI with the puported inadequacy of Lisp, note that Brooks is a Lisp guy and at least his original subsumption work relied on the Scheme programming language (a variant of Lisp), perhaps implemented directly in a chip.
(Note that the poster I'm replying to didn't make these inferences -- this is just a convenient place for me to attach this info.)
I have an impression that according to Minsky mind is something completely authonomous, with its own absolute (logical rules), something you can implement in a machine (hence the whole notion of AI).
Actually, Minsky doesn't believe in 'logical rules'. The Stanford group (following McCarthy) believes that logic is the royal road to AI. For this, they are termed the 'neats' and contrasted with the more pragmatic 'scruffies' such as the folks under Minsky's direction at MIT. Incidentally, there are 'neats' at MIT too -- for example, Chomsky -- who are most definitely at odds with Minsky and his, well, Minions.
There is another point of view (which I believe is more adequate) which boils down to mind and intellect as fairly sophisticated adaptation function (or tool), so to implement AI we have to start with very simple machine capable of interacting with its environment, learning, adapting, and evolving.
What makes you think this is not AI? Logic, Minsky's work, Brooks's robots, genetic algorithms -- these are all approaches to AI. The domain of AI is not isomorphic with logic.
Minsky separates the mind from the body, from the environment, from the whole human exprience, and this is plainly wrong.
Well this is your opinion, as well as that of the 'embodied cognition' folks and other latter-day phenomenologists. It may or may not be correct. Frankly, the embodied approach has not produced the great AI breakthrough that has eluded classical approaches.
Perhaps all of these paradigmatic battles and half-truths are the poisonous work of philosophers? Nah, couldn't be.
Linux is good for those who value their freedom (speech and beer)
Your comment is tongue and cheek, I know, but just to be clear: Linux is for those who value the freedom to see the source -- that's it. There are other kinds of freedom, and I doubt you can make platform-dependent generalizations for these.
Back to the article: Of course such comparisons are meaningless, but I found it interesting that they gave the Interface comparison to Windows (by a nose). Their argument was essentially 'we know the Windows interface, it remains unchanged, therefore it's better'. By this logic, OS X will only win when it becomes more familiar (read: Windows-like). Weird.