But you're fighting a straw man, because basically nobody in AI or AGI is trying to use vocabularies of shapes, or geometry in the simple sense you describe, to do vision processing...
You foolishly believe that the things you subjectively perceive as "shapes" cannot be recognized & comprehended by algorithmic processes... simply because you, personally, lack the education and/or mental capability to figure out how to do such a thing... But if society were limited to the science and tech Mike Tintner can understand, we'd still be in the Stone Age, painting shapes upon the cave walls ;) ben g On Thu, Jun 7, 2012 at 2:14 PM, Mike Tintner <[email protected]>wrote: > ** > Ben: It seems you persistently confuse the subjective feeling that humans > have when conducting an activity, with the underlying mechanisms that > enable them to perform the activity. > > No, as with patchworks, you're not prepared to make the mental leap > necessary to look at the world in a different way. > > To complete my line of argument - the brain's alphabet of basic forms can > be seen as comprising a "vocabulary" of body **shapes**. Art is primarily > about shapes. In fact, as the art of Arcimboldo, Chuck Close & others > demonstrate, we often actually don't need the component forms of those > shapes to recognize them. Faces can be painted of component forms like > fruit, plates, and geometric figures - and we can still recognize them from > their basic shapes. > > Dance and physical movement and embodied understanding of other bodies are > directed primarily by shapes. We dance primarily by having an image of the > shape we wish to achieve. We understand others' body movements by "putting > ourself into the position(/SHAPE)" of their bodies. We start from shapes, > not from a part-by-part, organ-by-organ, muscle-by-muscle analysis. You > have an idea of the shape of body movement you want to aim for, and that is > what informs your actual dance movements. Ditto we understand how Matisse's > Dancers dance with no information about their bodies other than their > outline shapes, and from those shapes alone we can get up and dance like > them. > > So when I talk about the ART OF MOVEMENT, I am talking primarily about > the shapes and shaping of movement. Art is about real world shapes. > Geometry is only a means for "analysing" shapes into artificial regular > components like triangles and circles. But it does not offer a vocabulary > of real world body shapes as art does. And it does not even offer a > sub-vocabulary of real world components, like hearts, livers, nerves, > neuronal networks, hands, fingers, toes. > > Successful AGI robots will also think in terms of shapes. It's possible - > you have to use imagination and be prepared to think differently. > > I'm trying to explain why AGI is getting nowhere - and point to the new > directions it must take. You're trying to defend the indefensible old > directions. > > *From:* Ben Goertzel <[email protected]> > *Sent:* Thursday, June 07, 2012 6:19 PM > *To:* AGI <[email protected]> > *Subject:* Re: [agi] The Visual Alphabet > > Mike T, > > >> The "so what" is this - while I take your point, Ben, I doubt that any >> AGI exercises, a la Poggio & Arel, are actually going to be any better - >> once you allow for the fact that AGI-tests are always on much more heavily >> controlled & selected/ "configured" materials. >> > > I understand your feelings, but the Google results are irrelevant to the > success of AGI methods. > > That's like saying the Wright Brothers were doomed to fail, based on the > evidence that Icarus-type flight methods failed ;p .... The failure of one > sort of method, does not doom all other approaches.... If you see all > different AI / AGI methods as basically the same, that's just because you > have terrible vision ... > > Do you have any tests that show any real AGI promise whatsoever? >> > > No tests that would impress skeptics, let alone an anti-rational ideologue > such as yourself ;) > > Similarly, before the Wright Brothers first flight, they had no test > results that would have impressed skeptics or anti-flight ideologues. > Flight skeptics sure wouldn't have given a rat's ass about the wind tunnel > tests they did while perfecting their wings... > > > >> >> The thesis I wish to play with is that just as words reduce to letters, >> so do visual forms reduce to a basic alphabet of abstract forms (albeit not >> a precise, closed alphabet) - which would include such basic units as the >> blob, wedge, chunk, (irregular), and the circle, triangle, square, >> (regular) and straight and zig-zag or wavy lines. >> >> And a Google or more AGI exercise of recognition may be able to, with >> some mild but unimpressive success, identify similar basic forms. >> > > There is a fairly obscure approach to computer vision, called image > grammars, that works as you're suggesting. Indeed, its results are pretty > poor so far. > > Google's image processing works based on totally unrelated methods, that > use statistical and machine learning methods to find highly complex > mathematical combinations of large numbers of very low-level image > features (much lower level than the "basic units" you mention) that imitate > human image categorization judgments. > > Your fascination with these topics is admirable, but you repeatedly > demonstrate that you lack the technical background to really understand the > things you're talking about ;-p ... > > **** > P.S. I was looking for a phrase as I wrote this & I may have got it - > it's not enough to know about the physics and biomechanics of movement, in > order to understand bodies - you have to know about the ART OF MOVEMENT. > Human and animal movement are indeed an art - both in terms of production > and the final forms of movement. Knowing about biomechanics will not help > you understand or predict the infinitely open-ended range of dances that > humans keep producing. Understanding and creating those dances is > literally, technically, an art/art form - just as drawing dancers is. > *** > > This kind of argumentation is very uninteresting at the current stage, > because chess masters would also argue that chess is an art form too. > Also, I feel like doing integrals in calculus is an art form, but > Mathematica does them better than me. Basically, whatever AI hasn't > conquered yet, will be called a "art form that computers can't emulate" -- > until AI has conquered it, and then it gets crossed off the list ;p > > It seems you persistently confuse the subjective feeling that humans have > when conducting an activity, with the underlying mechanisms that enable > them to perform the activity. > > For example, if we fall in love, we may feel that our beloved is the most > beautiful and amazing person in the world. But that doesn't make it > objectively true. > > Similarly, we may feel that we are carrying out some amazing > non-computational intuitive process when we move our pawn on the chess > board, or raise our arm while playing tennis But that doesn't make it so, > from a physical and scientific perspective. > > These are elementary philosophical errors you're making, but you seem to > be stuck in a rut of sorts, and maybe you won't ever get out, unless > someone forcibly modifies your brain. I have a strange feeling that, even > if you were shown an AGI that could do everything a human can do, you still > wouldn't believe it was **really** intelligent, and would just consider it > an irrelevant parlor trick ;) ... > > -- Ben G > > > *AGI* | Archives <https://www.listbox.com/member/archive/303/=now> > <https://www.listbox.com/member/archive/rss/303/6952829-59a2eca5> | > Modify<https://www.listbox.com/member/?&;>Your Subscription > <http://www.listbox.com> > *AGI* | Archives <https://www.listbox.com/member/archive/303/=now> > <https://www.listbox.com/member/archive/rss/303/212726-11ac2389> | > Modify<https://www.listbox.com/member/?&;>Your Subscription > <http://www.listbox.com> > -- Ben Goertzel, PhD http://goertzel.org "My humanity is a constant self-overcoming" -- Friedrich Nietzsche ------------------------------------------- AGI Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-c97d2393 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-2484a968 Powered by Listbox: http://www.listbox.com
