>"How many bloody times are you guys going to make the same mistake over and over? It seems to be an almost universal error here.>Maths is formulaic – it seeks to produce the formulae for SETS of abstract forms. But it can only produce formulae for sets of REGULAR forms."
*Todor: *That's wrong, I explained you why it seems so to you. The "regular forms" are just the simplest ones you see. Maths and physics includes very complex differential equations - as complex as you wish, - namely laws to solve those. *>"Until you get this distinction into your head – maths CAN deal with INDIVIDUAL irregular forms, but CAN’T deal with SETS OR FAMILIES of irregular forms, >you are going to have a major problem understanding AGI – in fact be *incapable* of understanding AGI.* *>The central task for AGI is to visually recognize/ conceptualise/ generate FAMILES and GROUPS of IRREGULAR FORMS, not just indiividual ones..* *."* *Todor: *Maths is about generalization of rules, not about individual forms. The individual forms are instances and application of those rules. But I agree that AGI should be capable to cope with "irregular forms" if that's interpreted as: to accumulate complexity and to build more complex "forms" than the basic ones (the "regular" ones). *>Maths can represent any INDIVIDUAL rock form. It cannot produce a math formula for a SET of rock forms – for rock forms generally* *Todor: *In fact maths and physics are created exactly to deal with the "forms" (functions, correlations, phenomenons) seen in nature (sensory inputs), and exactly with sets (generalizations) which match particular requirements/limits, classes of things. Even the basics in maths - the numbers - are in fact "sets" and the formulas are exactly a way to apply rules to a class or families of objects with similar characteristics. The "irregular" forms are just forms which require more complex set of parameters and "irregular" rules to get defined. *>"This is what defeats all attempts to visually object recognize and conceptualise. Our computers cannot recognize human faces and bodies because they are so >irregular. Nor can they produce any conceptual prototypes for such forms.""* *Todor: *I don't think so, and faces are not irregular. The nonessential parameters are "irregular", that's why they are non-essential, why they look "irregular and complex" and why they cannot be generalized (it's a web like the causation). Only the repeating features are generalized. IMHO there are practical facial and body recognition and 3D pose estimation tools, there are not used widely yet, though, but it is about to get massively popular very soon as a part of the "Natural Interfaces". In fact one case of 3D pose estimation (human bodies) is getting popular already, Kinect and other sensors like Xite, which are RGB+Depth - the direct depth sensing makes 3D reconstruction easier, when there are skeleton models for humans pre-built (like in Kinect), it's even more easier. Cheers -- *--- Todor "Tosh" Arnaudov ---* * -- Twenkid Research:* http://research.twenkid.com -- *Self-Improving General Intelligence Conference*: http://artificial-mind.blogspot.com/2012/07/news-sigi-2012-1-first-sigi-agi.html *-- Todor Arnaudov's Researches Blog**: *http://artificial-mind.blogspot.com ------------------------------------------- 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
