John Rose said: "The paper is about engineering play a priori into systems since play is involved with self-organization, attractors, emergent universality, decision instability, and play is inherent in many things even neurons and octopi. Perhaps even wiggling amoebas are having a party "
Well that is a friendly commentary. I don't think that the author's basic idea is novel enough to generate anything really new, and his personal application of these kinds of ideas does not lead to any powerful insights about how we might solve the problems of AI/AGI or about the nature of emergent systems in general. So play may be an emergent mechanism...The same thing could be said about learning in general. But why didn't the author get to that point where he realized that it is learning that takes place in play and (therefore it would be learning that helps shape the brain? Because he, at least at some intuitive level, understands that other kinds of learning may involve directed inquiry. (And that other kind of learning does take place during play.) "Is play some biomechanical attempt to modify causal entropic force on the edge of chaos," John Rose asks. No, (general) causal entropic force is not a material event that can be modified by an animal and it cannot be defined on the basis of sensory events alone. So the answer has to be no. Perhaps I am quibbling. I think that point of view makes sense. Play is a mechanism that helps an animal to modify causal entropic forces. It is just that it is neither a workable explanation nor is it a powerful enough explanation of the processes that go on during play and learning in general that would make it profoundly true. Jim Bromer On Sat, Dec 13, 2014 at 8:54 AM, John Rose via AGI <[email protected]> wrote: > The paper is about engineering play a priori into systems since play is > involved with self-organization, attractors, emergent universality, > decision instability, and play is inherent in many things even neurons and > octopi. Perhaps even wiggling amoebas are having a party J > > > > Then, is play in an organism some biomechanical attempt to modify causal > entropic force on the edge of chaos? Like in baseball trying to hit a single > verses hitting a foul into the bleachers. Fouling is chaos. Hitting it in > play is order. > > > > John > > > > From: Jim Bromer via AGI [mailto:[email protected]] > Sent: Saturday, December 13, 2014 8:10 AM > > > To: AGI > Subject: Re: [agi] Physics, Emergence, and the Connectome > > > > I am not really sure what he means because the polarity (or magnetic > orientation) of a 'permanent magnet' can be reversed if an external magnetic > field is strong enough. I think there are a few things wrong with his > presentation. The sense that learning is not part of a computational theory > of mind is a little far-fetched but then he seems to present play as an > unexpected explanation without any mention of the word "learn" (other than > in the introduction where he says, "We speak instead about learning more > about the connections themselves and about advancing medicine (Van Essen and > Ugurbil, 2012xVan Essen, D.C. and Ugurbil, K. Neuroimage. 2012; 62: > 1299–1310 > > CrossRef | PubMed | Scopus (54)See all ReferencesVan Essen and Ugurbil, > 2012).") It seems like a major oversight. > > > Jim Bromer > > > > On Sat, Dec 13, 2014 at 8:00 AM, John Rose via AGI <[email protected]> wrote: > > He means the properties of some things are the same everywhere, under > similar conditions, like a water molecule’s molecular dynamics on Earth > would be the same on another planet one million years from now… or.. mostly > the same there might be slight subatomic differences. But you know what he > means. > > > > I’m wondering, does play occur on the edge of chaos? And maybe on edges of > chaos? > > > > John > > > > From: Jim Bromer via AGI [mailto:[email protected]] > Sent: Saturday, December 13, 2014 6:29 AM > To: AGI > Subject: Re: [agi] Physics, Emergence, and the Connectome > > > > "The rigidity and elasticity of all solids is always the same. The special > properties of insulators, semiconductors, and metals that allow us to make > computers with them are always the same. The rigid orientation of > ferromagnets is always the same." > > > > Whaa...? > > > Jim Bromer > > > > On Sat, Dec 13, 2014 at 3:16 AM, Russ Hurlbut via AGI <[email protected]> > wrote: > > Here's an interesting paper that touches on some of the recurring themes > discussed on the AGI list: > > > > http://www.cell.com/neuron/fulltext/S0896-6273(14)00681-3 > > > > For those that care to weigh in, perhaps the perspective in this open access > article can provide another facet to reflect on familiar topics including > complexity, compression, and the paper's central theme - play. > > AGI | Archives | Modify Your Subscription > > AGI | Archives | Modify Your Subscription > > > > AGI | Archives | Modify Your Subscription ------------------------------------------- AGI Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-f452e424 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657 Powered by Listbox: http://www.listbox.com
