Re: [agi] Tommy
> Also anything you can find on case-based reasoning, tho it is woefully rare. Having done a lot of case-based reasoning almost 23 years ago . . . . Case-based reasoning is effectively analogous to weighted nearest neighbor in multi-dimensional space. If you (or the system) can define the dimensions and scale and weight them, it's an awesome method -- this is equivalent to the logic-based/expert-system approach to CBR. The other alternative, which most people don't realize is exactly equivalent to CBR, is to just use neural networks (since they just effectively "map" the multi-d space -- complete with scaling and weighting). Having used both methods, I would say that, until they both scale themselves fairly quickly into oblivion, the neural network method is more accurate while CBR provides much better explanations. The unfortunate thing is that as you add more and more dimensions, both methods falter pretty quickly. - Original Message - From: "J Storrs Hall, PhD" <[EMAIL PROTECTED]> To: Sent: Monday, May 14, 2007 7:51 AM Subject: Re: [agi] Tommy > On Saturday 12 May 2007 10:24:03 pm Lukasz Stafiniak wrote: > >> Do you have some interesting links about imitation? I've found these, >> not all of them interesting, I'm just showing what I have: > > Thanks -- some of those look interesting. I don't have any good links, but > I'd > reccomend Hurley & Chater, eds, Perspectives on Imitation (in 2 vols). > > Also anything you can find on case-based reasoning, tho it is woefully rare. > > Josh > > - > This list is sponsored by AGIRI: http://www.agiri.org/email > To unsubscribe or change your options, please go to: > http://v2.listbox.com/member/?&; > - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
On Saturday 12 May 2007 10:24:03 pm Lukasz Stafiniak wrote: > Do you have some interesting links about imitation? I've found these, > not all of them interesting, I'm just showing what I have: Thanks -- some of those look interesting. I don't have any good links, but I'd reccomend Hurley & Chater, eds, Perspectives on Imitation (in 2 vols). Also anything you can find on case-based reasoning, tho it is woefully rare. Josh - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
On Sunday 13 May 2007 08:14:43 am Kingma, D.P. wrote: > John, as I wrote earlier, I'm very interested in learning more about your > particular approach to: > - Concept and pattern representation. (i.e. types of concept, patterns, > relations?) As I mentioned in another note, (about the tennis ball), a concept is a set of programs that embody your abilities to recognize, manipulate, and predict some thing, and to make inferences about its past. > - Concept creation. (searching for statistically signifcant spatiotemporal > correlations, genetic programming, neural networks, ?) I think that the brain has a lot of what Ben calls "pattern mining" in the hardware; and for our purposes, conventional (including some recent) work in ML/PR seems to be adequate. The main key to the process, though, is to have what programming language theorists call a "reflective tower": for the language to be able to represent itself, and thus reason about its own programs, and thus about the programs it uses to reason about programs, etc, etc. (BTW, this is where standard programming language theory fails us, in that they are very enamored of provably complete logics below the Gödel line. This is all very well for writing provably correct red-black tree implementations, but they are playing with pebbles on the beach, as it were, with a great ocean of truth lying undiscovered before them.) > - Concept revision/optimization. You mention you use search techniques, > could you be a little more specific? I try to represent the data as numeric vectors as much as possible so I can use standard scientific regression methods to create the functions that predict it. Discrete stuff is harder -- AI has tackled it for 50 years with modest results. However, it's my intuition and hope that continuous models underneath (and lots of brute force processing) will provide the "intuition" to conquer the combinatorial explosion. (Consider backprop: the finished neural net is not unlike the same function implemented in perceptrons, except that with a step function instead of a sigmoid you have no traction whatsoever for hill-climbing towards an optimum.) > Unfortunately you did not go into specifics yet. Since you wrote that your > ideas are "firm enough to start doing experiments", I was hoping you could > give a glimpse of the underlying idea's. I.e., what's your "You're speaking > of a "high-level functional reactive programming language" exactly? "If I were going to spend 8 hours chopping down a tree," said Abraham Lincoln, "I'd spend the first 6 sharpening my axe." When you're trying to write an AI, you don't need the distractions and extra work of worrying about storage allocation, process migration and communication, data formats, and so forth -- those wheels have already been invented (as have statistical analysis, regression, and so forth). Build a system with those easily available and interoperable, and worry about AI things thereafter. The easiest languages to "reflect" about are the functional ones -- ones where the semantics resemble math more than assembly language. This is all well and good, except that the pure functional paradigm leaves out the ability to deal with time. (You can't write X=X+1 in a functional language, because it isn't true!) There are 4 main ways people have tried to extend functional languages to deal with time. The first is ad hoc, as in Lisp: mix a functional language with an imperative one. Second, in what were called "applicative state transition" systems, model a big state machine and use the functional language to write the transition function. Third, the current favorite in the PLT crowd, is category theoretic monads: write a function that computes a list that is a trace of the behavior you want the program to enact. And finally, functional reactive programming, is to write a function that is interpreted as a circuit, where each value is actually a signal that varies in time. This, it turns out, is how the physicists and engineers have done it all along: a systems-and-signals circuit in control theory or cybernetics is isomorphic to a FR program. > One reason I'm interested is that there are many approaches to unsupervised > learning of physical (or just graphical) concepts, and I'm thinking of Serre > et al, Hawkins, neural network specialists (Geoffrey Hinton) and there could > be many more. But none of these theories I know of are strong enough to > extract high-level concepts such as 'gravity'. Neither is the average human being -- Newton was one of the great geniuses. What the average human does is expect things to fall down. Experiments with people who haven't studied physics show that they have a very poor ability to predict what will happen in simple experiments. (For example, more than half of undergrads and non-science faculty, when asked to draw the path that would be taken by a ball rolling off the edge of a table, don't draw anything
Re: [agi] Tommy
John, as I wrote earlier, I'm very interested in learning more about your particular approach to: - Concept and pattern representation. (i.e. types of concept, patterns, relations?) - Concept creation. (searching for statistically signifcant spatiotemporal correlations, genetic programming, neural networks, ?) - Concept revision/optimization. You mention you use search techniques, could you be a little more specific? Unfortunately you did not go into specifics yet. Since you wrote that your ideas are "firm enough to start doing experiments", I was hoping you could give a glimpse of the underlying idea's. I.e., what's your "You're speaking of a "high-level functional reactive programming language" exactly? One reason I'm interested is that there are many approaches to unsupervised learning of physical (or just graphical) concepts, and I'm thinking of Serre et al, Hawkins, neural network specialists (Geoffrey Hinton) and there could be many more. But none of these theories I know of are strong enough to extract high-level concepts such as 'gravity'. Now when I imagine a hypothetical system capable of extracting such thing as 'gravity', he would have to go through a process of many stages, one of the first to learn about the general spatial phenomenon 'object', later the temporal phenomenon 'velocity', and would eventually find out that the vertical element of the velocity vector is ever decreasing with constant amount. Now, a system that is capable of doing this without any prior knowledge is pretty damn interesting, and you're promising it. I've been doing some vaguely similar experiments (minus motor feedback) with a multilayer spatio-temporal pattern classifier network and concluded that it's not easy (in contrary) to let a system extract concepts like 'gravity'. Kind regards, Durk Kingma On 5/12/07, J Storrs Hall, PhD <[EMAIL PROTECTED]> wrote:On Friday 11 May 2007 08:26:03 pm Pei Wang wrote: As you can see from my comment and paper, I agree with your idea in > its basic spirit. However, I think your above presentation is too > vague, and far from enough for semantic analysis. True enough -- my ideas tend to form like planets, a la the nebular hypothesis :-) But at this point, having gnawed on them for a few years, I think they're firm enough to start doing experiments. On 5/13/07, J Storrs Hall, PhD <[EMAIL PROTECTED]> wrote: On Saturday 12 May 2007 09:00:46 am Pei Wang wrote: > I see --- it is fine to stress the procedural aspect of concept given > your context. However, to make your design flexible and general, even > in that case you will still need some "language" to specify your > concepts, rather than in a pin-ball-specific manner, right? Sure -- tho in my case it looks more like a very high-level functional reactive programming language than FOPL. Josh - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
On 5/13/07, J Storrs Hall, PhD <[EMAIL PROTECTED]> wrote: On Saturday 12 May 2007 09:00:46 am Pei Wang wrote: > ...My understanding is that ..., your "world model" is, in essence, a bunch of "if I > do this, I'll observe that", which is a summary of experience, or > interactions between the system and its environment, rather than the > environment "by itself". It's both. As Yogi Berra said, "You can observe a lot by just watching." So the model is a bunch of "This happened, and thus that happened", where "I did this" is a particularly important special case of "this happened". But watching someone else do something is key to imitation, which is key to learning. Do you have some interesting links about imitation? I've found these, not all of them interesting, I'm just showing what I have: * [[Learning How to Do Things with Imitation -> http://citeseer.ist.psu.edu/339624.html]] * [[Reinforcement Learning with Imitation in Heterogeneous Multi-Agent Systems -> http://citeseer.ist.psu.edu/35684.html]] * [[http://citeseer.ist.psu.edu/jenkins00primitivebased.html | Primitive-Based Movement Classification for Humanoid Imitation]] * [[Self-Segmentation of Sequences: Automatic Formation of Hierarchies of Sequential Behaviors -> http://citeseer.ist.psu.edu/286643.html]] * [[Imitation as a First Step to Social Learning in Synthetic Characters: A Graph-based Approach -> http://alumni.media.mit.edu/~daphna/sca_final_electronic.pdf]] * [[Human's Meta-cognitive Capacities and Endogenization of Mimetic Rules in Multi-Agents Models -> http://www.uni-koblenz.de/~essa/ESSA2003/ChavalariasESSA03.pdf]] * [[http://girardianlectionary.net/covr2004/Chavalariasabst.pdf | Metareflexive Mimetism: The prisoner free of the dilemma]] * [[http://www.aisb.org.uk/publications/proceedings/aisb05/3_Imitation_Final.pdf | Proceedings of the Third International Symposium on Imitation in Animals and Artifacts]] * [[http://ecagents.istc.cnr.it/dllink.php?id=214&type=Document | The progress drive hypothesis: an interpretation of early imitation]] - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
On Saturday 12 May 2007 09:00:46 am Pei Wang wrote: > ...My understanding is that ..., your "world model" is, in essence, a bunch of "if I > do this, I'll observe that", which is a summary of experience, or > interactions between the system and its environment, rather than the > environment "by itself". It's both. As Yogi Berra said, "You can observe a lot by just watching." So the model is a bunch of "This happened, and thus that happened", where "I did this" is a particularly important special case of "this happened". But watching someone else do something is key to imitation, which is key to learning. > I see --- it is fine to stress the procedural aspect of concept given > your context. However, to make your design flexible and general, even > in that case you will still need some "language" to specify your > concepts, rather than in a pin-ball-specific manner, right? Sure -- tho in my case it looks more like a very high-level functional reactive programming language than FOPL. Josh - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
Thanks! I'll be in touch. Josh On Saturday 12 May 2007 10:08:26 am Derek Zahn wrote: > > > [EMAIL PROTECTED] writes: > > Help from anyone on this list with experience with the GNU toolchain on > ARM-based microcontrollers will be gratefully accepted :-) > > I have a lot of such experience and would be happy to help out with whatever you need. Post more details here if you think they are of general interest or otherwise just contact me directly: [EMAIL PROTECTED] . - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
On Saturday 12 May 2007 09:18:16 am Mike Tintner wrote: > Josh:My major hobby-horse in this area is that a concept has to be an active > machine, capable of recognition, generation, inference, and prediction. > > This sounds very like Jeff Hawkins, (just reading On Intelligence now). Do > you see your position as generally accepted, or at the forefront of changing > AI attitudes to concepts? Procedural embedding of knowledge was used, and the phrase introduced, by Winograd in the 1970's. It became passe in the 80s when people tried to pack lots of "knowledge" into the expert systems but essentially traded quality for quantity. BTW, Hawkins has been discussed at length here -- some of his ideas are valuable, but none is particularly original, and in many places where he says things like "nobody has tried or is doing X" he's often speaking from ignorance. > And if it's not too much to ask (and it may be), would you care to give a > particular concept example of what you mean? Consider my concept of a tennis ball. I have circuitry in my brain -- a neural FPGA is closer to the way I think about it than a sequential program is -- to recognize it when I see it, when I feel it, when I hear it bounce, when I put my foot down on it without having seen it. I have similar machinery for throwing it, and for predicting what it's going to do when it's thrown by someone else. Indeed the circuitry is good enough to control a racquet within the seconds of arc angle and milliseconds of time to volley it to a chosen spot when it's hit at me at over 50 FPS -- these circuits are specialized enough that you can tell from an EEG trace whether a tennis pro is playing with natural or synthetic strings in his racquet, so it's pretty clear that they are specific to tennis balls as the projectile. My concept of a tennis ball includes the ability to squeeze one and vary my predictions of its trajectory after a bounce depending on the feel. It includes the motor circuitry to shape the hand to hold 2, 3, or 4 of them (not easy) and to know how many I have in my pocket by the pressure on the leg and the stretch of the pants fabric. Of course it also includes declarative stuff like the facts that they are yellow (but were typically white 30 years ago), round, 2.25" in diameter, and cost about a dollar. Josh - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
RE: [agi] Tommy
> [EMAIL PROTECTED] writes: > Help from anyone on this list with experience with the GNU toolchain on > > ARM-based microcontrollers will be gratefully accepted :-) I have a lot of such experience and would be happy to help out with whatever you need. Post more details here if you think they are of general interest or otherwise just contact me directly: [EMAIL PROTECTED] . - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
Josh:My major hobby-horse in this area is that a concept has to be an active machine, capable of recognition, generation, inference, and prediction. This sounds very like Jeff Hawkins, (just reading On Intelligence now). Do you see your position as generally accepted, or at the forefront of changing AI attitudes to concepts? And if it's not too much to ask (and it may be), would you care to give a particular concept example of what you mean? - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
On 5/12/07, J Storrs Hall, PhD <[EMAIL PROTECTED]> wrote: On Friday 11 May 2007 08:26:03 pm Pei Wang wrote: > *. Meaning come from experience, and is grounded in experience. I agree with this in practice but I don't think it's necessarily, definitionally true. In practice, experience is the only good way we know of to build the models that provide us the ability to predict the world. AI tried it by hand-building models throughout the 80s (the "expert system" era) and mostly failed. However, if I have a new robot, I can copy the bits from an old one and its mind will have just as much meaning as the old one. Thus in theory, any other way I could have come up with the same string of bits will also give me meaning. That is also my plan. "Experience" is not restricted to direct, personal experience. When I'm reading, I'm getting other people's experience. The key difference is that whether the meaning of a concept is determined by its experienced relation with others concepts, or by its "denotation" in the world. Model-theoretic semantics in logic has a meaning more or less opposite that of the use of "model" in AI -- in the former case the world is a "model" for the logical system, in the latter the logical system is a model of the world. In that sense, yes, but even in AI, "meaning" is still traditionally treated as denotation, that is, the outside object/event referred to by a symbol. If you want your robot to build a "world model" to describe the world "as it is", it will run into the same trouble as model-theoretic semantics. My understanding is that this is not what you mean. Instead, your "world model" is, in essence, a bunch of "if I do this, I'll observe that", which is a summary of experience, or interactions between the system and its environment, rather than the environment "by itself". > I fully agree with your focus. I guess your "concepts" are patterns or > structures formed from certain "semantic primitives" by a fixed set of > operators or connectors. I'm very interested in your choice. My major hobby-horse in this area is that a concept has to be an active machine, capable of recognition, generation, inference, and prediction. Of course we know that any machine can be represented by a program and thus given a "declarative" representation, but for practical purposes, I'm fairly far over toward the "procedural embedding of knowledge" end of the spectrum. I see --- it is fine to stress the procedural aspect of concept given your context. However, to make your design flexible and general, even in that case you will still need some "language" to specify your concepts, rather than in a pin-ball-specific manner, right? Pei - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
That's a cute comparison, but pretty insulting both to infants and developmental psychology, which continues to paint an ever more detailed picture of what restless, exploratory scientists infants are. Damn noisy too, I agree. Surely a major challenge for AGI/robotics is to build an agent/robot that is only fractionally as exploratory. - Original Message - From: "J Storrs Hall, PhD" <[EMAIL PROTECTED]> To: Sent: Saturday, May 12, 2007 12:41 PM Subject: Re: [agi] Tommy On Friday 11 May 2007 08:55:12 pm Mike Tintner wrote: ...All these machines you are talking about are basically inert lumps of metal and don't exist without human beings to switch them on, feed them & interpret them. Same is true of a baby, except for the part where you can turn it off and get a full night's sleep. Josh - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?&; -- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.5.467 / Virus Database: 269.6.8/800 - Release Date: 11/05/2007 19:34 - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
The point re the computer-human nexus is simply that it's rather like the horse doing the trick of counting for its master - if the master is there, you can't be sure that the horse really is doing counting or understands what it's doing, or that it could do the trick without its master. Let me give you another analogy if it's of any help. The whole of science in applying its current mechanistic paradigm to the world has also forgotten that machines don't exist without humans. So cognitive psychology treats the human mind like a computer. Actually it would be much truer and more productive to treat the human mind like a human-computer hybrid, i.e. a human using a computer. That opens new dimensions on human thinking - you realise that human intelligent performance may not be so much a case of some people having, and others not having, certain faculties but of some people using, and others not using their faculties - as some do and some don't use their computer's faculties (something that scientific psychology rarely considers). So I disagree - you have to look at the totality of how your computer is used by a human, before you can be sure of how intelligent it is or isn't. And I'm not trying to be difficult or patronising - if the whole of science and major scientific minds can leave out the human factor, so can Pei. - Original Message - From: "Pei Wang" <[EMAIL PROTECTED]> To: Sent: Saturday, May 12, 2007 12:10 PM Subject: Re: [agi] Tommy Mike, I just wonder that whenever you find "there's one thing so screamingly obvious that you guys don't seem to be taking it into account", have it ever occurred to you that there may be a valid reason? For the current issue, whether there is still human "in the loop" has little to do with the machine's intelligence, as far as the human is not responsible for specifying the machine's operation step-by-step. In the long run, machines will surely become more and more autonomous, but we probably still want to stay in the loop, even though technically it won't be necessary. Anyway, "taking humans out of the loop" doesn't sound like a good choice for AGI development at the moment, unless you have a concrete design to show us otherwise. Pei On 5/11/07, Mike Tintner <[EMAIL PROTECTED]> wrote: > Josh, > Since the 90s there has been a strand in AI research that claims that >> robotics is necessary to the enterprise, based on the notion that >> having a body is necessary to intelligence. Symbols, it is said, must >> be grounded in physical experience to have meaning. Without such >> grounding AI practitioners are deceiving themselves by calling their >> Lisp atoms things like MOTHER-IN-LAW when they really mean no more >> than G2250. > Pei: I think these people correctly recognized a problem in traditional AI, > though they attributed it to a wrong cause.. Every implemented system > already has a "body" --- the hardware, and as long as the system has input and output, it has experience that comes from its body. Of course, since the body is not human body, the experience is not human experience. However, as far as this discussion is concerned, it doesn't matter, since this kind of experience is genuine experience that can be used to ground meaning of concepts. Er, there's one thing so screamingly obvious that you guys don't seem to be taking it into account here. All these machines you are talking about are basically inert lumps of metal and don't exist without human beings to switch them on, feed them & interpret them. Humans are still, pace Rodney B, "in the loop." Try taking humans out of the loop and then see what these standalone computers do and don't understand - or what they do, period. - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?&; - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?&; -- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.5.467 / Virus Database: 269.6.8/800 - Release Date: 11/05/2007 19:34 - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
On Friday 11 May 2007 08:26:03 pm Pei Wang wrote: > *. Meaning come from experience, and is grounded in experience. I agree with this in practice but I don't think it's necessarily, definitionally true. In practice, experience is the only good way we know of to build the models that provide us the ability to predict the world. AI tried it by hand-building models throughout the 80s (the "expert system" era) and mostly failed. However, if I have a new robot, I can copy the bits from an old one and its mind will have just as much meaning as the old one. Thus in theory, any other way I could have come up with the same string of bits will also give me meaning. > A more detained discussion and a proposed solution can be found in > http://nars.wang.googlepages.com/wang.semantics.pdf Model-theoretic semantics in logic has a meaning more or less opposite that of the use of "model" in AI -- in the former case the world is a "model" for the logical system, in the latter the logical system is a model of the world. To avoid any confusion, let me point out that I always use the word in the AI sense. > As you can see from my comment and paper, I agree with your idea in > its basic spirit. However, I think your above presentation is too > vague, and far from enough for semantic analysis. True enough -- my ideas tend to form like planets, a la the nebular hypothesis :-) But at this point, having gnawed on them for a few years, I think they're firm enough to start doing experiments. > > 2. The hard part is learning: the AI has to build its own world > > model. My instinct and experience to date tell me that this is > > computationally expensive, involving search and the solution of > > tough optimization problems. > > Agree, though I've been avoiding the phrase "world model", because the > intuitive picture it provides: there is a "objective world" out there, > and an AI is building an "internal model" of it, where the concepts > represent objects, and beliefs represent factual relations among > objects --- this is a picture you don't subscribe, I guess. "World model" has a very well established meaning in AI (50 years old by now) and I find the basic idea sound. I DON'T think that one should assume at the outset there are objects and relations -- I'm using a representation where objects can be represented if experience indicates it's a useful category, but other ways of representing the world are equally accessible. > A good idea. As I said above: input/output is necessary for AGI, but > any concrete form of them is not, in principle. An AGI doesn't have to > be able to move itself around in the physical world (though it must > somehow change its environment), and doesn't have to have a certain > human sensor (though it must somehow sense its environment). Agreed. > I'd suggest to add the "muscle" in as soon as possible to get a > complete sensor-motor cycle. Help from anyone on this list with experience with the GNU toolchain on ARM-based microcontrollers will be gratefully accepted :-) > I fully agree with your focus. I guess your "concepts" are patterns or > structures formed from certain "semantic primitives" by a fixed set of > operators or connectors. I'm very interested in your choice. My major hobby-horse in this area is that a concept has to be an active machine, capable of recognition, generation, inference, and prediction. Of course we know that any machine can be represented by a program and thus given a "declarative" representation, but for practical purposes, I'm fairly far over toward the "procedural embedding of knowledge" end of the spectrum. > > I claim that most current AI experiments that try to mine meaning out > > of experience are making an odd mistake: looking at sources that are > > too rich, such as natural language text found on the Internet. The > > reason is that text is already a highly compressed form of data; it > > takes a very sophisticated system to produce or interpret. Watching a > > ball roll around a blank tabletop and realizing that it always moves > > in parabolas is the opposite: the input channel is very low-entropy > > (in actual information compared to nominal bits), and thus there is > > lots of elbow room for even poor, early, suboptimal interpretations to > > get some traction. > > I don't think you have convinced me that this kind of experiment is > better than the others (such as those in NLP) , but you get a good > idea and it is worth a try. “Two roads diverged in a yellow wood and I, I took the path less travelled by, and that has made all the difference.” Josh - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
On Friday 11 May 2007 09:15:56 pm Mike Tintner wrote: > I'm saying the last 400 years have been framed by Descartes' and science's > mind VERSUS body dichotomy. That in turn has been expressed in a whole > variety of subsidiary dichotomoies and cultural battles: > ... mind vs body > ... reason vs emotion ... I think all your comparisons are actually different axes in reality, and they don't have enough in common to conflate usefully. For example, in my reading of the Scientific/Industrial Revolution, body and reason got accelerated while mind and emotion got left behind. Couldn't happen if they were the same dichotomy. Josh - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
On Friday 11 May 2007 08:55:12 pm Mike Tintner wrote: >...All these machines you are talking about are > basically inert lumps of metal and don't exist without human beings to > switch them on, feed them & interpret them. Same is true of a baby, except for the part where you can turn it off and get a full night's sleep. Josh - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
Mike, I just wonder that whenever you find "there's one thing so screamingly obvious that you guys don't seem to be taking it into account", have it ever occurred to you that there may be a valid reason? For the current issue, whether there is still human "in the loop" has little to do with the machine's intelligence, as far as the human is not responsible for specifying the machine's operation step-by-step. In the long run, machines will surely become more and more autonomous, but we probably still want to stay in the loop, even though technically it won't be necessary. Anyway, "taking humans out of the loop" doesn't sound like a good choice for AGI development at the moment, unless you have a concrete design to show us otherwise. Pei On 5/11/07, Mike Tintner <[EMAIL PROTECTED]> wrote: > Josh, > Since the 90s there has been a strand in AI research that claims that >> robotics is necessary to the enterprise, based on the notion that >> having a body is necessary to intelligence. Symbols, it is said, must >> be grounded in physical experience to have meaning. Without such >> grounding AI practitioners are deceiving themselves by calling their >> Lisp atoms things like MOTHER-IN-LAW when they really mean no more >> than G2250. > Pei: I think these people correctly recognized a problem in traditional AI, > though they attributed it to a wrong cause.. Every implemented system > already has a "body" --- the hardware, and as long as the system has input and output, it has experience that comes from its body. Of course, since the body is not human body, the experience is not human experience. However, as far as this discussion is concerned, it doesn't matter, since this kind of experience is genuine experience that can be used to ground meaning of concepts. Er, there's one thing so screamingly obvious that you guys don't seem to be taking it into account here. All these machines you are talking about are basically inert lumps of metal and don't exist without human beings to switch them on, feed them & interpret them. Humans are still, pace Rodney B, "in the loop." Try taking humans out of the loop and then see what these standalone computers do and don't understand - or what they do, period. - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?&; - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
Computational AI/AGI vsrobotics (Symbolic AIvs (situated, embodied evolutionary robotics) What has been happening over the last decade or so, is that all these dichotomies have been dissolving. It's arguably a consensus now that you can't have reason without emotion, but the other dichotomies and battles are still raging including throughout AI. It is true that these dichotomies are still a subject of debate among AI academics, but I actually think a significant percentage of people on this list agree that they are misleading dichotomies, and that you can potentially "have your cake and eat it too" by integrating symbolic methods with low-level perception/action stuff as occurs in real and simulated robotics. Certainly, Novamente incorporates this kind of integration in its design principles... and it is not the only AGI design to do so... -- Ben - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
Josh, [ignore previous truncated version] I'm not quite sure what your angle is here, but I don't seem to be communicating, (please correct me). If BTW you and/or others aren't interested in this whole cultural history area, please ignore. I'm saying the last 400 years have been framed by Descartes' and science's mind VERSUS body dichotomy. That in turn has been expressed in a whole variety of subsidiary dichotomoies and cultural battles: HUMAN SYSTEM: mind vs body Self vs body reason vs emotion rationality vs imagination intelligence vs creativity (convergent vs divergent intelligence intelligence) logic vs analogy intellect vs athleticism SIGN SYSTEMS/ MEDIA literacy vs artistic education (symbolic vs image media media ) (language, vs painting, photography, video etc) maths ORGANIZED KNOWLEDGE cognitive psychology vs physical psychology cognitive sciences physiological psychology embodied cognition sciences vs arts (general particular, abstract concrete) [science vs religion] philosophy vs naturalistic, science-based "other-wordly," philosophy thought-experimental AI Computational AI/AGI vsrobotics (Symbolic AIvs (situated, embodied evolutionary robotics) What has been happening over the last decade or so, is that all these dichotomies have been dissolving. It's arguably a consensus now that you can't have reason without emotion, but the other dichotomies and battles are still raging including throughout AI. Very soon now, I'm arguing, there will be a consensus about all these things - and in every case, it will be recognised that you can't have the left side, the pure, rational, disembodied, symbolic side WITHOUT the right side, without the imaginative, emotional, imagistic side - can't have mind without body - or AGI without a robotic body. You will have a corporate science and AI that sees them all as inseparable sides of a whole. All this is happening now - dualism may have been bankrupt a while ago, but Dennett has been, and still is, spending a massive amount of energy arguing against it, because its influence is still playing out, including in the current battles of AI.. What Josh: On Friday 11 May 2007 03:06:52 pm Mike Tintner wrote: ... the mind/body era inaugurated by Descartes (& the first scientific revolution) is coming to an end right across our culture? Dualism was intellectually bankrupt by 1950, with the spate of mechanized logic results from Godel, Turing, Church, Kleene, etc, and Shannon's information theory and Weiner & Rosenblueth's "Teleology" paper that was one of the foundations of cybernetics. The illusion of pure, ethereal, rational mind, which takes so many forms, is fast fading. The scientific revolution was informed by the notion that the physical world was mechanistic and worked by laws that could be written down and understood. Descartes' dualism was a step TOWARD that from the earlier assumption that both mind and body were moved by mystical "life forces." Dualism said that only the mind was, the body was mechanistic. The intellectual revolution of the 20th century merely dropped the other shoe, saying that both mind and body are mechanistic. The revolution in physical movers was back that-a-way -- somewhere in the Midlands they should be celebrating the 300th anniversary of the beginning of the Industrial Revolution right about now. It's been a while since putting a motor in a car earned it the name "auto-mobile" -- nowadays we take for granted that it can move by itself, and use "auto" to mean things that control themselves as well. The era of mind -- mechanical rather than ethereal -- is just beginning. Josh - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?&; -- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.5.467 / Virus Database: 269.6.8/797 - Release Date: 10/05/2007 17:10 - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
Josh, I'm not quite sure what your angle is here, but I don't seem to be communicating, (please correct me). If BTW you and/or others aren't interested in this whole cultural history area, please ignore. I'm saying the last 400 years have been framed by Descartes' and science's mind VERSUS body dichotomy. That in turn has been expressed in a whole variety of subsidiary dichotomoies and cultural battles: HUMAN SYSTEM: mind vs body Self vs body reason vs emotion rationality vs imagination intelligence vs creativity (convergent vs divergent intelligence intelligence) logic vs analogy SIGN SYSTEMS/ MEDIA literacy vs artistic education (symbolic vs image media media ) What Josh: On Friday 11 May 2007 03:06:52 pm Mike Tintner wrote: ... the mind/body era inaugurated by Descartes (& the first scientific revolution) is coming to an end right across our culture? Dualism was intellectually bankrupt by 1950, with the spate of mechanized logic results from Godel, Turing, Church, Kleene, etc, and Shannon's information theory and Weiner & Rosenblueth's "Teleology" paper that was one of the foundations of cybernetics. The illusion of pure, ethereal, rational mind, which takes so many forms, is fast fading. The scientific revolution was informed by the notion that the physical world was mechanistic and worked by laws that could be written down and understood. Descartes' dualism was a step TOWARD that from the earlier assumption that both mind and body were moved by mystical "life forces." Dualism said that only the mind was, the body was mechanistic. The intellectual revolution of the 20th century merely dropped the other shoe, saying that both mind and body are mechanistic. The revolution in physical movers was back that-a-way -- somewhere in the Midlands they should be celebrating the 300th anniversary of the beginning of the Industrial Revolution right about now. It's been a while since putting a motor in a car earned it the name "auto-mobile" -- nowadays we take for granted that it can move by itself, and use "auto" to mean things that control themselves as well. The era of mind -- mechanical rather than ethereal -- is just beginning. Josh - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?&; -- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.5.467 / Virus Database: 269.6.8/797 - Release Date: 10/05/2007 17:10 - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
Josh, Since the 90s there has been a strand in AI research that claims that robotics is necessary to the enterprise, based on the notion that having a body is necessary to intelligence. Symbols, it is said, must be grounded in physical experience to have meaning. Without such grounding AI practitioners are deceiving themselves by calling their Lisp atoms things like MOTHER-IN-LAW when they really mean no more than G2250. Pei: I think these people correctly recognized a problem in traditional AI, though they attributed it to a wrong cause.. Every implemented system already has a "body" --- the hardware, and as long as the system has input and output, it has experience that comes from its body. Of course, since the body is not human body, the experience is not human experience. However, as far as this discussion is concerned, it doesn't matter, since this kind of experience is genuine experience that can be used to ground meaning of concepts. Er, there's one thing so screamingly obvious that you guys don't seem to be taking it into account here. All these machines you are talking about are basically inert lumps of metal and don't exist without human beings to switch them on, feed them & interpret them. Humans are still, pace Rodney B, "in the loop." Try taking humans out of the loop and then see what these standalone computers do and don't understand - or what they do, period. - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
Josh, This is an interesting idea that deserves detailed discussion. Since the 90s there has been a strand in AI research that claims that robotics is necessary to the enterprise, based on the notion that having a body is necessary to intelligence. Symbols, it is said, must be grounded in physical experience to have meaning. Without such grounding AI practitioners are deceiving themselves by calling their Lisp atoms things like MOTHER-IN-LAW when they really mean no more than G2250. I think these people correctly recognized a problem in traditional AI, though they attributed it to a wrong cause. My opinion on this issue can be summarized as the following: *. Meaning come from experience, and is grounded in experience. *. However, for AGI, this "experience" doesn't have to be "human experience". *. Every implemented system already has a "body" --- the hardware, and as long as the system has input and output, it has experience that comes from its body. Of course, since the body is not human body, the experience is not human experience. However, as far as this discussion is concerned, it doesn't matter, since this kind of experience is genuine experience that can be used to ground meaning of concepts. *. The failure of traditional AI is not to use standard computer hardware rather than special hardware (i.e., robot), but to ignore the experience of the system when handling meaning of concepts. A more detained discussion and a proposed solution can be found in http://nars.wang.googlepages.com/wang.semantics.pdf This has given rise to a plethora of silly little robots (in Minsky's view, anyway) that scurry around the floor picking up coffeecups and like activities. I also think it is not a fruitful direction for AI to move. My view lies somewhere between the extremes on this issue: a) Meaning does not lie in a physical connection. I find meaning in the concept of price-theoretical market equilibria; I've never seen, felt, or smelled one. Meaning lies in working computational models, and true meaning lies in ones that ones that can make correct predictions. As you can see from my comment and paper, I agree with your idea in its basic spirit. However, I think your above presentation is too vague, and far from enough for semantic analysis. b) On the other hand, the following are true: 1. Without some connection to external constraints, there is a strong temptation on the part of researchers to define away the hard parts of the AI problem. Even with the best will in the world, this happens subconsciously. Agree. 2. The hard part is learning: the AI has to build its own world model. My instinct and experience to date tell me that this is computationally expensive, involving search and the solution of tough optimization problems. Agree, though I've been avoiding the phrase "world model", because the intuitive picture it provides: there is a "objective world" out there, and an AI is building an "internal model" of it, where the concepts represent objects, and beliefs represent factual relations among objects --- this is a picture you don't subscribe, I guess. "That deaf, dumb, and blind kid sure plays a mean pinball." Thus Tommy. My robotics project discards a major component of robotics that is apparently dear to the embodiment crowd: Tommy is stationary and not autonomous. This not only saves a lot of construction but allows me to run the AI on the biggest system I can afford (currently ten processors) rather than having to shoehorn code and data into something run off a battery. A good idea. As I said above: input/output is necessary for AGI, but any concrete form of them is not, in principle. An AGI doesn't have to be able to move itself around in the physical world (though it must somehow change its environment), and doesn't have to have a certain human sensor (though it must somehow sense its environment). Tommy, the pinball wizard kid, was chosen as a name for the system because of a compelling, to me anyway, parallel between a pinball game and William James' famous description of a baby's world as a "blooming, buzzing confusion." The pinball player is in the same position as a baby in that he has a firehose input stream of sensation from the lights and bells of the game, but can do little but wave his arms and legs (flip the flippers), which very rarely has any effect at all. Makes sense. Tommy, the robot, consists at the moment of a pair of Firewire cameras and the ability to display messages on the screen and receive keyboard input -- ironically almost the exact opposite of the rock opera Tommy. Planned for the relatively near future is exactly one "muscle:" a single flipper. Tommy's world will not be a full-fledged pinball game, but simply a tilted table with the flipper at the bottom. I'd suggest to add the "muscle" in as soon as possible to get a complete sensor-motor cycle. Tommy, the scientific experiment and engineering proj
Re: [agi] Tommy
On 11/05/07, J Storrs Hall, PhD <[EMAIL PROTECTED]> wrote: Tommy, the scientific experiment and engineering project, is almost all about concept formation. He gets a voluminous input stream but is required to parse it into coherent concepts (e.g. objects, positions, velocities, etc). None of these concepts is he given originally. Tommy 1.0 will simply watch the world and try to imagine what happens next. Interesting. This is somewhat similar to one of the projects that I am interested in. Assuming sufficient or the correct hardware, I'm interested in body mounted robotics for Intelligence Augmentation, using what people would think of as AI. An example of the robot if not the software http://www.robots.ox.ac.uk/ActiveVision/Projects/Wear/wear.03/index.html I would start off with that annotating its visual streams to be passed to head mounted display on the user. Things like tracking objects the user has pointed at, so the user could see things not directly in front of him, or high-lighting important objects to the user, would be some of the things it would be initially taught. I would also give it a controlled, low power laser pointer so it could visually mark things for other people apart from its user. I think this sort of system is a worthy one to study, as it allows the user and the robot to inhabit the same world (so concepts developed by the computer should not be too alien to the user, and thus languages may be shared between them), it also allows for long periods of time for the researcher to be present with the computer if such time scales as a babies development are required for the teaching of human level intelligence. It also tries to minimise the amount of processing/robotics required to share the similar world, meaning more projects could possibly be attempted at once. While user and computer do share the same world in your experimental setup, there may be some concepts that would be hard for it to learn such as translation of its PoV. Whether that would be a fatal flaw in its developed mental model of the world (and limit its ability to communicate with as hardware and its capabilities developed), I'm not sure. More experimentation and better theories required, as ever. Will Pearson - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
On Friday 11 May 2007 03:06:52 pm Mike Tintner wrote: >... the mind/body era inaugurated by > Descartes (& the first scientific revolution) is coming to an end right > across our culture? Dualism was intellectually bankrupt by 1950, with the spate of mechanized logic results from Godel, Turing, Church, Kleene, etc, and Shannon's information theory and Weiner & Rosenblueth's "Teleology" paper that was one of the foundations of cybernetics. > The illusion of pure, ethereal, rational mind, which > takes so many forms, is fast fading. The scientific revolution was informed by the notion that the physical world was mechanistic and worked by laws that could be written down and understood. Descartes' dualism was a step TOWARD that from the earlier assumption that both mind and body were moved by mystical "life forces." Dualism said that only the mind was, the body was mechanistic. The intellectual revolution of the 20th century merely dropped the other shoe, saying that both mind and body are mechanistic. The revolution in physical movers was back that-a-way -- somewhere in the Midlands they should be celebrating the 300th anniversary of the beginning of the Industrial Revolution right about now. It's been a while since putting a motor in a car earned it the name "auto-mobile" -- nowadays we take for granted that it can move by itself, and use "auto" to mean things that control themselves as well. The era of mind -- mechanical rather than ethereal -- is just beginning. Josh - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
MT;In the final and the first analysis, the brain is a device for controlling movement Josh: Only half, even in a hunter/gatherer context. The other half is participation in the social process, which in its essence is pure communication. The social point/dimension is true, as covered in a previous thread. But re "pure" communication, don't you have an awareness - having written this book, (which sorry I haven't read) - that the mind/body era inaugurated by Descartes (& the first scientific revolution) is coming to an end right across our culture? The illusion of pure, ethereal, rational mind, which takes so many forms, is fast fading. And, actually, the autonomous mobile robot will symbolize and define the new era even more powerfully than the clock defined the first? (In fact, historically I understand, it was the then hydraulic ROBOTS in the royal gardens that actually inspired Descartes' mechanism just as much as the clock). P.S. Is the term for tenure, then, rigor academici? - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
Yes, thank you, a meaningful and very interesting project. I discussed this kind of system with a friend of mine half an hour ago. On 5/11/07, J Storrs Hall, PhD <[EMAIL PROTECTED]> wrote: 2. The hard part is learning: the AI has to build its own world model. My instinct and experience to date tell me that this is computationally expensive, involving search and the solution of tough optimization problems. This must be the central part of your project. I'm very interested in how you approach the following problems: - Concept and pattern representation. If you use some sort of graphical model, what types of edges, nodes, relations? Something like Ben's SMEPH? - Concept creation. Do you have single method in mind or multiple methods, maybe working simultaneously? Data mining methods, statistical methods, genetic programming, NN's (e.g. Boltzmann machines), ...? - Concept revision/optimization. You mention you use search techniques, could you be a little more specific (or references). Will there be something like a wake/sleep cycle, or is optimization done in real-time? Also, why did you choose a physical implementation and not a virtual one? Simply because it's more interesting or are there other motives? These kind of project are, of course, very complex and multi-faceted, but worth is because they force you to think about these extremely essential things like model creation, concept formation, model optimization. (BTW I ordered your new book "Beyond AI" this week, and looking forward to reading it.). Please keep us updated on your project. Kind regards, Durk Kingma "That deaf, dumb, and blind kid sure plays a mean pinball." Thus Tommy. My robotics project discards a major component of robotics that is apparently dear to the embodiment crowd: Tommy is stationary and not autonomous. This not only saves a lot of construction but allows me to run the AI on the biggest system I can afford (currently ten processors) rather than having to shoehorn code and data into something run off a battery. Tommy, the pinball wizard kid, was chosen as a name for the system because of a compelling, to me anyway, parallel between a pinball game and William James' famous description of a baby's world as a "blooming, buzzing confusion." The pinball player is in the same position as a baby in that he has a firehose input stream of sensation from the lights and bells of the game, but can do little but wave his arms and legs (flip the flippers), which very rarely has any effect at all. Tommy, the robot, consists at the moment of a pair of Firewire cameras and the ability to display messages on the screen and receive keyboard input -- ironically almost the exact opposite of the rock opera Tommy. Planned for the relatively near future is exactly one "muscle:" a single flipper. Tommy's world will not be a full-fledged pinball game, but simply a tilted table with the flipper at the bottom. Tommy, the scientific experiment and engineering project, is almost all about concept formation. He gets a voluminous input stream but is required to parse it into coherent concepts (e.g. objects, positions, velocities, etc). None of these concepts is he given originally. Tommy 1.0 will simply watch the world and try to imagine what happens next. The scientific rationale for this is that visual and motor skills arrive before verbal ones both in ontogeny and phylogeny. Thus I assume they are more basic and the substrate on which the higher cognitive abilities are based. Furthermore I have a good idea what concepts need to be formed for competence in this area, and so I'll have a decent chance of being able to tell if the system is going in the right direction. I claim that most current AI experiments that try to mine meaning out of experience are making an odd mistake: looking at sources that are too rich, such as natural language text found on the Internet. The reason is that text is already a highly compressed form of data; it takes a very sophisticated system to produce or interpret. Watching a ball roll around a blank tabletop and realizing that it always moves in parabolas is the opposite: the input channel is very low-entropy (in actual information compared to nominal bits), and thus there is lots of elbow room for even poor, early, suboptimal interpretations to get some traction. Josh - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?&; - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
Right. The key issue is autogeny in the mental architecture. Learning will be unsupervised to start, with internal feedback from how well the system is expecting what it sees next. Then we move into a mode where imitation is the key, with the system trying to do what a person just did (e.g. catching the ball on the flipper, hitting some certain spot on the table, etc (note the flipper control is a full 1-dof signal, not just a 1-bit button). I can catch a ping-pong ball on a paddle in 3-d -- Tommy should be able to learn it in 2-d with an effective 0.1 G field!) To do this he'll have to develop concepts to describe what it is I'm trying to do. There's a LOT you can do with 1 DOF output -- you could even imagine Tommy passing the Turing Test by sending Morse code with the flipper :-) Josh On Friday 11 May 2007 01:52:31 pm Derek Zahn wrote: > Bob Mottram writes:> In order to differentiate this from the rest of the robotics crowd you> need to avoid building a specialised pinball playing robot. > > I can't speak for JoSH, but I got the impression that playing "pinball" or anything similar was not the object, the object was to provide real sensor data in a somewhat limited domain to experiment with and observe concept formation. You'd like to see it develop object permanence, ball motion, gravity, "bouncing", and so on. The goal not being so much to impress people with performance on a vertical task but rather to use the task environment as a somewhat rich sandbox in which general purpose capabilities can be studied. - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
Friday, May 11, 2007, J Storrs Hall, PhD wrote: JSHP> 2. The hard part is learning: the AI has to build its own world JSHP> model. And for this it requires complex enough world to model. Information about the world can be given by static description (which also includes action-reaction pairs, but doesn't depend on system's actions), or dynamically, providing data on complex requests of the system. Physical embodiment provides means to access world by interaction (dynamic description). Static description of physical world (as it can be accessed in 'narural' ways through vision, hearing, etc.) is not dense in interesting patters and is extremely expensive to analyze. If you limit interaction with world to that single flipper, it won't change situation dramatically from static description. And static description can be given in much mode dense way using some form of NL-based code. -- Vladimir Nesovmailto:[EMAIL PROTECTED] - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
On Friday 11 May 2007 02:01:09 pm Mike Tintner wrote: ... > As Daniel Wolpert will tell you, the sea squirt devours its brain as soon as > it stops moving. As Dan Dennet has pointed out, this resembles what happens when one gets tenure... > In the final and the first analysis, the brain is a device > for controlling movement: Only half, even in a hunter/gatherer context. The other half is participation in the social process, which in its essence is pure communication. Manipulation of the physical world remains important but has declined relative to communication significantly in the modern world. > (And, thinking aloud as I write, ALL senses are moving senses. Animals, > including the simplest one-celled organisms, move their sensors around to > perceive the world. Touch too, of course - you have to move your body to > get a hold of things ). Ultimately I'm thinking in terms of a Cog-like torso with hands -- but that's many years off. > P.P.S. Can't resist this - set your robot free:: > I'm Free > [TOMMY] Nice. Maybe someday... Josh - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
Josh, Interesting work, and I like the nature of your approach. We have essentially a kind of a pin ball machine at IDSIA and some of the guys were going to work on watching this and trying to learn simple concepts from the observations. I don't work on it so I'm not sure what the current state of their work is. When you publish something on this please let the list know! thanks Shane - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
Josh: Thus Tommy. My robotics project discards a major component of robotics that is apparently dear to the embodiment crowd: Tommy is stationary and not autonomous As Daniel Wolpert will tell you, the sea squirt devours its brain as soon as it stops moving. In the final and the first analysis, the brain is a device for controlling movement: "Movement is the only way we have of interacting with the world, whether foraging for food or attracting a waiter's attention. Indeed, all communication, including speech, sign language, gestures and writing, is mediated via the motor system. Taking this viewpoint, the purpose of the human brain is to use sensory signals to determine future actions. The goal of our lab is to understand the computational principles underlying human sensorimotor control." http://learning.eng.cam.ac.uk/wolpert/ Computational and Biological Learning Lab Having written the book, so to speak, aren't you best placed to know that this is the age of autonomous MOBILE robots? P.S. The other interesting thing here is that evolutionarily, touch precedes vision, no? And the two, I suggest, are intertwined in a brain that works by "common sense" rather than isolated senses. Michael Tye, I think, (I forget the name of his theory), has pointed out that we have the illusion that we can isolate our senses - just see things, for example - whereas in fact our sensory perception of the world is always a common sense one. (And, thinking aloud as I write, ALL senses are moving senses. Animals, including the simplest one-celled organisms, move their sensors around to perceive the world. Touch too, of course - you have to move your body to get a hold of things ). P.P.S. Can't resist this - set your robot free:: I'm Free [TOMMY] I'm free -- I'm free, And freedom tastes of reality! I'm free -- I'm free, And I'm waiting for you to follow me. If I told you what it takes To reach the highest high, You'd laugh and say "Nothing's that simple." But you've been told many times before Messiahs pointed to the door And no one had the guts to leave the temple! I'm free -- I'm free, And I'm waiting for you to follow me. I'm free -- I'm free, And I'm waiting for you to follow me. - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
RE: [agi] Tommy
Bob Mottram writes:> In order to differentiate this from the rest of the robotics crowd you> need to avoid building a specialised pinball playing robot. I can't speak for JoSH, but I got the impression that playing "pinball" or anything similar was not the object, the object was to provide real sensor data in a somewhat limited domain to experiment with and observe concept formation. You'd like to see it develop object permanence, ball motion, gravity, "bouncing", and so on. The goal not being so much to impress people with performance on a vertical task but rather to use the task environment as a somewhat rich sandbox in which general purpose capabilities can be studied. - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
Re: [agi] Tommy
In order to differentiate this from the rest of the robotics crowd you need to avoid building a specialised pinball playing robot. If the machine can learn and form concepts based upon its experiences it should be able to do so with any kind of game, provided that suitable actuators are attached. It is very easy to fall into the trap of building something which is just a physical expert system. From long experience of trying to do things like that I think there is no getting around the fact that in order to be truly general you have to build world models upon which reasoning systems can act, which means getting into the tricky business of modelling sensors and probabilistic interactions. It is possible to take much simpler Brooksian approaches, but in these cases what you always end up with is a brittle expert system. This might be ok if all you're trying to do is model insect-like intelligence operating within some well defined niche, but ideally we want our robots to be smarter than cockroaches. On 11/05/07, J Storrs Hall, PhD <[EMAIL PROTECTED]> wrote: Since the 90s there has been a strand in AI research that claims that robotics is necessary to the enterprise, based on the notion that having a body is necessary to intelligence. Symbols, it is said, must be grounded in physical experience to have meaning. Without such grounding AI practitioners are deceiving themselves by calling their Lisp atoms things like MOTHER-IN-LAW when they really mean no more than G2250. This has given rise to a plethora of silly little robots (in Minsky's view, anyway) that scurry around the floor picking up coffeecups and like activities. My view lies somewhere between the extremes on this issue: a) Meaning does not lie in a physical connection. I find meaning in the concept of price-theoretical market equilibria; I've never seen, felt, or smelled one. Meaning lies in working computational models, and true meaning lies in ones that ones that can make correct predictions. b) On the other hand, the following are true: 1. Without some connection to external constraints, there is a strong temptation on the part of researchers to define away the hard parts of the AI problem. Even with the best will in the world, this happens subconsciously. 2. The hard part is learning: the AI has to build its own world model. My instinct and experience to date tell me that this is computationally expensive, involving search and the solution of tough optimization problems. "That deaf, dumb, and blind kid sure plays a mean pinball." Thus Tommy. My robotics project discards a major component of robotics that is apparently dear to the embodiment crowd: Tommy is stationary and not autonomous. This not only saves a lot of construction but allows me to run the AI on the biggest system I can afford (currently ten processors) rather than having to shoehorn code and data into something run off a battery. Tommy, the pinball wizard kid, was chosen as a name for the system because of a compelling, to me anyway, parallel between a pinball game and William James' famous description of a baby's world as a "blooming, buzzing confusion." The pinball player is in the same position as a baby in that he has a firehose input stream of sensation from the lights and bells of the game, but can do little but wave his arms and legs (flip the flippers), which very rarely has any effect at all. Tommy, the robot, consists at the moment of a pair of Firewire cameras and the ability to display messages on the screen and receive keyboard input -- ironically almost the exact opposite of the rock opera Tommy. Planned for the relatively near future is exactly one "muscle:" a single flipper. Tommy's world will not be a full-fledged pinball game, but simply a tilted table with the flipper at the bottom. Tommy, the scientific experiment and engineering project, is almost all about concept formation. He gets a voluminous input stream but is required to parse it into coherent concepts (e.g. objects, positions, velocities, etc). None of these concepts is he given originally. Tommy 1.0 will simply watch the world and try to imagine what happens next. The scientific rationale for this is that visual and motor skills arrive before verbal ones both in ontogeny and phylogeny. Thus I assume they are more basic and the substrate on which the higher cognitive abilities are based. Furthermore I have a good idea what concepts need to be formed for competence in this area, and so I'll have a decent chance of being able to tell if the system is going in the right direction. I claim that most current AI experiments that try to mine meaning out of experience are making an odd mistake: looking at sources that are too rich, such as natural language text found on the Internet. The reason is that text is already a highly compressed form of data; it takes a very sophisticated system to produce or interpret. Watching a ball roll around a blank
RE: [agi] Tommy
J. Storrs Hall writes: > Tommy, the scientific experiment and engineering project, is almost> all > about concept formation. Great project! While I'm not quite sure about "meaning in the concept of price-theoretical market equilibria" thing, I really like your idea and it's similar in broad concept to my as yet very early noodling. A couple of comments: * To the casual observer "Tommy" implies that your AI is blind, deaf, and dumb, which might not quite be the idea you are trying to convey. * It would seem more robust, easier, and cooler to pick up a real used pinball machine and use it instead of the abstract idealized pinball machine. I look forward to seeing some results and asking: "How do you think he does it? I don't know! What makes him so good?" - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936