Re: [agi] analogy, blending, and creativity
On 6/2/07, J Storrs Hall, PhD [EMAIL PROTECTED] wrote: And many scientists refer to potential energy surfaces and the like. There's a core of enormous representational capability with quite a few well-developed intellectual tools. Another Grand Unification theory: Estimation of Distribution Algorithms behind Bayesian Nets, Genetic Programming and unsupervised Neural Networks. - 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=231415user_secret=e9e40a7e
Re: [agi] analogy, blending, and creativity
On 5/17/07, J Storrs Hall, PhD [EMAIL PROTECTED] wrote: On Wednesday 16 May 2007 04:47:53 pm Mike Tintner wrote: Josh . If you'd read the archives, you'd see that I've advocated constructive solid geometry in Hilbert spaces as the basic representational primitive. Would you like to say more re your representational primitives? Sounds interesting. The archives have no reference to constructive solid geometry in Hilbert spaces in any form. Personally, I think it's a plot. MOOO ha ha ha! It's all in your mind :-) Actually, I can't find it either but (and this is apropos to the subject) we rarely remember the exact words we said or heard; we remember more abstract representations. Chances I used CSG and/or vector spaces. Hilbert space is a rhetorical flourish anyway -- they may need it to describe quantum mechanics precisely but we'll never implement it... Many engineering departments make the mistake of never mentioning the term Hilbert space and calling it all signal analysis. - 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=231415user_secret=e9e40a7e
Re: [agi] analogy, blending, and creativity
And many scientists refer to potential energy surfaces and the like. There's a core of enormous representational capability with quite a few well-developed intellectual tools. Josh On Saturday 02 June 2007 08:31:07 am Lukasz Stafiniak wrote: On 5/17/07, J Storrs Hall, PhD [EMAIL PROTECTED] wrote: On Wednesday 16 May 2007 04:47:53 pm Mike Tintner wrote: Josh . If you'd read the archives, you'd see that I've advocated constructive solid geometry in Hilbert spaces as the basic representational primitive. Would you like to say more re your representational primitives? Sounds interesting. The archives have no reference to constructive solid geometry in Hilbert spaces in any form. Personally, I think it's a plot. MOOO ha ha ha! It's all in your mind :-) Actually, I can't find it either but (and this is apropos to the subject) we rarely remember the exact words we said or heard; we remember more abstract representations. Chances I used CSG and/or vector spaces. Hilbert space is a rhetorical flourish anyway -- they may need it to describe quantum mechanics precisely but we'll never implement it... Many engineering departments make the mistake of never mentioning the term Hilbert space and calling it all signal analysis. - 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=231415user_secret=e9e40a7e
Re: [agi] analogy, blending, and creativity
Josh : A blend is more like designing a helicopter by combining a dragonfly and a car. You take the general shape and behavior of the dragonfly, and the size, interior seats, driver controls, etc, from a car. In general in a blend you start with B and C without an A. Both relations B-D and C-D are analogies, mappings from one concept to another. (You could, if you wanted, complete the square (quadrature) and find an A -- but it's typically the opposite of what you want. In the example you'd get a beetle, with the size and provenance of the dragonfly and the shape and behavior of the car!) Josh, Good example of how analogy/ blending works. But then you seem to be going back (in my terms) to the idea that analogy works by symbolic mappings. No way, I would suggest, that your example - or indeed the overwhelming percentage of analogy - works in that fashion. (If you or anyone else disagrees, please construct even the beginning of a symbolic mapping for your example). The obvious term that I have been searching for in these discussions is MORPHING. The way analogy actually works, I suggest, both in your example and most of the time, is by the brain morphing one graphic/image into another or into a composite.. Since the animal/human brain is continually seeing shapes morphing all the time - zooming in or out - as they move towards or away from the viewer - morphing is fundamental to perception. And, of course, the brain is continually morphing in dreams - constructing/morphing new, never-before-experienced shapes out of old, actually-experienced ones. That's what imagination is overwhelmingly - morphing. ReSHAPING the world, not, for the most part, reWORDING it. P.S. We need a new term for the symbol addiction (and resistance to image-ination) of AI and our literate culture generally. It's a very serious, hard-to-kick addiction! - 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=231415user_secret=fabd7936
Re: [agi] analogy, blending, and creativity
John, Thanks for your reply to my questions about your project Tommy in your previous post. I'm very interested about the details but please forgive my relative freshness to this field (CS graduate heading to an AI master :) I'm particularly interested in the types of pattern mining you're planning to perform, the types of patterns you think are necessary to search for (spatial, temporal, causal, etc), and which search techniques you're going to use given time constraints. But I guess you will release more information as the project continues. Mike Tintner wrote on May 16, 2007 3:13 PM: Good example of how analogy/ blending works. But then you seem to be going back (in my terms) to the idea that analogy works by symbolic mappings. As far as I can see, John is not referring to any particular symbolic implementation, but to analogy in general. You can go both ways: NARS and Copycat both find analogies, the former symbolically but the latter is somewhere between symbolic and connectionist. The obvious term that I have been searching for in these discussions is MORPHING. The way analogy actually works, I suggest, both in your example and most of the time, is by the brain morphing one graphic/image into another or into a composite.. The brain does not use images for representation, except tiny patches in the very 'lowest' regions in the visual cortex. Representation is abstract, distributed. You could read Seeing and Visualizing: It's Not What You Think by Zenon W. Pylyshyn for a comprehensive synthesis of research and theory. Your idea that some kind of 'morphing' happens in the brain is not new. An interesting technique (imho) is Geoffrey Hinton's RBM (Restricted Boltzmann Machine) which is a form of generative neural network. After training on handwriting digits it can perform 'confabulation' which means that the network wanders between different consistent constraint states. This is results in interesting 'movies', which look like simplified versions of human imagination and dream. You should see them, it's fairly consistent with your view of 'morphing'. Despite the fairly limited amount of artificial neurons and the networks' generative nature, they perform very well (last time I checked, the best) on the MNIST handwritten digit database set. It is computionally expensive though. - 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=231415user_secret=fabd7936
Re: [agi] analogy, blending, and creativity
On Wednesday 16 May 2007 09:13:27 am Mike Tintner wrote: Good example of how analogy/ blending works. But then you seem to be going back (in my terms) to the idea that analogy works by symbolic mappings. Seem in this case is in the eye of the beholder. If you'd read the archives, you'd see that I've advocated constructive solid geometry in Hilbert spaces as the basic representational primitive. 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=231415user_secret=fabd7936
Re: [agi] analogy, blending, and creativity
Josh . If you'd read the archives, you'd see that I've advocated constructive solid geometry in Hilbert spaces as the basic representational primitive. Would you like to say more re your representational primitives? Sounds interesting. The archives have no reference to constructive solid geometry in Hilbert spaces in any form. Personally, I think it's a plot. - 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=231415user_secret=fabd7936
Re: [agi] analogy, blending, and creativity
On Wednesday 16 May 2007 04:47:53 pm Mike Tintner wrote: Josh . If you'd read the archives, you'd see that I've advocated constructive solid geometry in Hilbert spaces as the basic representational primitive. Would you like to say more re your representational primitives? Sounds interesting. The archives have no reference to constructive solid geometry in Hilbert spaces in any form. Personally, I think it's a plot. MOOO ha ha ha! It's all in your mind :-) Actually, I can't find it either but (and this is apropos to the subject) we rarely remember the exact words we said or heard; we remember more abstract representations. Chances I used CSG and/or vector spaces. Hilbert space is a rhetorical flourish anyway -- they may need it to describe quantum mechanics precisely but we'll never implement it... The basic idea is that after all, it's straightforward to represent the sensory inputs as a numeric vector ( one number for the signal on each nerve). Ditto for motor outputs. It's straightforward to represent the various transforms done by the visual and auditory systems as mappings in vector spaces. Furthermore, physical science has used numeric vectors as a basic form of representation for centuries. Together with the calculus, this is really the language of science. It seems silly to tackle the biggest scientific problem yet by throwing it away. So as a working hypothesis, I'm assuming that I can model the entire cognitive process with vector spaces / transforms as a basic representation, and that things like logic and language will show up as special cases that in some cases can be optimized (but shouldn't be until we understand what gets lost in the translation). CSG comes in when you have a space in which points represent something like frames, i.e. a description of a situation. Propositions can be represented as regions in the space where a given statement is true, and the mapping from propositional connectives to CSG operators is trivial. 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=231415user_secret=fabd7936
Re: [agi] analogy, blending, and creativity
On Wednesday 16 May 2007 11:14:57 am Kingma, D.P. wrote: John, BTW, it's Josh... I'm particularly interested in the types of pattern mining you're planning to perform, the types of patterns you think are necessary to search for (spatial, temporal, causal, etc), and which search techniques you're going to use given time constraints. But I guess you will release more information as the project continues. There is a fair stock of clustering and mining techniques, of which for a first crack at the problem I'm looking at affinity propagation and Kohonen maps. However, there is also the technique of having a set of pre-written transforms (e.g segmentation on pictures, Fourier transform on sound) that one can search for useful combinations of with a GA or the like. The brain does not use images for representation, except tiny patches in the very 'lowest' regions in the visual cortex. Representation is abstract, distributed. You could read Seeing and Visualizing: It's Not What You Think by Zenon W. Pylyshyn for a comprehensive synthesis of research and theory. Rem acu tetigisti! For those who think that we think in actual pictures, here's an exercise: imagine two ordinary gears, meshed, and turning. Let's say they are the same size and each one has 17 teeth. Got it? Okay, watch carefully as they turn. What shape are the cogs (teeth)? Do the cogs roll across each other's face or slide? What is the shape of the path taken by the contact point? Or, imagine a planetary gear setup. The sun gear has 12 cogs and there are three planet gears, each of which also has 12 cogs. How many cogs on the annular gear? Don't calculate, *count them* in your picture. Imagine the annulus is stationary, and the sun gear drives the mechanism so that each planet makes a complete revolution and returns to its original position. How many times did the sun gear turn? Don't calculate, *watch the picture.* Doesn't work, does it? Pylyshyn points out that the number of primitive trackers we have for objects in the visual field (he calls them FINSTs) is tiny, something like 4 or 5. We don't really see in complete pictures when we're looking at something, much less when we're imagining it. 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=231415user_secret=fabd7936