Hmmm... IMO, there was really nothing conceptually new in Hawkins' HTM, he just did a really good job of expressing some ideas that were already there in the literature. Which is an important and valuable thing, but doesn't really make his HTM model a good example of profound creativity.
To me a few better examples of creativity in science, would be, say, -- Darwin's discovery of natural selection -- Holland's creation of the "genetic algorithm" approach to optimization -- McCullough and Pitts' creation of the first "formal network model" (bridging the gap btw biology and computer science for the first time) Each of these actually introduced something new, rather than crystallizing and more clearly formulating ideas that were already there... Hawkins' idea of layering a Bayes net on top of a hierarchical neural net is creative, but not a really profound step IMO. I don't find it as significant as for instance -- Goldberg/Harik/Pelikan's fusion of probabilistic modeling w/evolutionary learning, resulting in the development of BOA and the launching of the field of Estimation of Distribution Algorithms Anyway I just don't "get" what you think is the "essence of the problem" that Novamente does not address. But you've discoursed on this theme at great length already and I still haven't "gotten it." I understand you think that analogizing based on sensorimotor experience, using internal simulation, is important, but this is a very old and well-known point which I and many other AGI researchers agree with you on ... -- Ben G On Nov 18, 2007 6:32 PM, Mike Tintner <[EMAIL PROTECTED]> wrote: > > > Ben: To be quite frank, the most creative and original ideas inside the > Novamente > design are quite technical. I suspect you don't have the background to > understand > them.However, I can see there's something else underlying your remarks: a > sort of > "mystification" of the power of intelligence. I think you'll be shocked > when the workings > of the brain are finally unveiled and ....no magic trick, no super-secret > "algorithm of thought" ... > > No, I don't have the background to judge MOSES etc. and, offhand say, the > novelty of your search algorithms. But I don't think that invalidates my > particular comments. And you offer a strange idea of creativity. Creativity > is not necessarily shocking. Clearly there was nothing shocking about > Hawkins' HTM. Creative ideas can often be like that - Freud's idea of the > unconscious was similar - involving a major new emphasis and development of > what is already known or broached. > > I certainly don't expect to see any super algorithms of thought. On the > contrary, I am very confident that human, conscious streams of thought are > *not* programmed or algorithmic overall, (while involving a great many > algorithmic routines in parts) - more "composition" in the ordinary sense > of throwing things together, than iteration of pre-existing instructions. > And that is the sort of revolution - among several others - that I expect > to see in both cog. sci and AGI/ robotics - revolutionary but not shocking > or magical. > > I certainly do not think that you do not have original ideas or that your > design is worthless - and I acknowledge, as before, your courage in pursuing > your project (and tolerance of criticism!) .But I also think it's no bad > thing that someone should keep complaining; "where's the beef?" and asking > that people address the central problem. And I think this discussion shows > that I am not alone. > > > ------------------------------ > 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=8660244&id_secret=66447870-126179
