AT, On Mon, May 4, 2015 at 1:51 PM, Anastasios Tsiolakidis < [email protected]> wrote:
> > On Mon, May 4, 2015 at 10:43 AM, Piaget Modeler <[email protected] > > wrote: > >> Does your adaptive control concept have anything to do with ACT-R >> (Adaptive Control of Thought - Rational) > > > > Sorry to jump in here but no, at least that's what my limited adaptive > engineering education tells me. But I do have a question about the > "adapters" here, how is adaptive control going to design a tool, ever? I > don't want to start lecturing again but it looks to me like tool > construction (more generally, science) depends on a) an instinct to > experiment, > Experimentation is fundamental to adaptive process control, as otherwise there would be no way to discover how to control anything. trial and error, > Part of experimentation. > fantasy and even wishful thinking > Leading to more experimetation. > b) on the compositionality of the world, which includes emergence: you can > break a stick or straw but not 10 or 1000 of them. This compositionality > essentially voids all statistical approaches, and I would include adaptive > control in statistics, essentially if someone started putting together > straw after straw an adaptive controller would start screaming louder and > louder "somebody stop that lunatic, it is an exercise in insanity". > Constant experimentation would stop such insanity. You seem to be concerned about exactly what I am concerned about, but seeing it from a different angle. Here are the two angles, and there are probably more: 1. What properties might a basic adapter need to be able to adapt all the way to intelligence? 2. Not seeing how to build basic adapters, how can we talk about intelligent systems made of them? I agree - we would have to solve this problem before going anywhere. However, watching the wonderful things that our own metabolic control systems do, it appears that there IS a solution. Steve ======= > > Having said that, and repeating myself somewhat, I think any "grounded" > research program is well worth it, even though the OP informs us that for > the time being he hasn't plugged-in "his" grounding. The particular EM bit > sounds somewhat arcane, the proof however remains in the pudding. Finally, > there is hardly anyone who wants new chip designs than me, obviously the > problem remains manufacturing and financing such manufacturing advances, > otherwise I am quite sure there lots of people who are looking into new > chip designs, including our friends and Cognimem, memristors etc. For my > part I am fooling around with an FPGA just to finetune my intuitions, but > can't imagine educating myself one way or the other all the way to chip > innovator. > > AT > *AGI* | Archives <https://www.listbox.com/member/archive/303/=now> > <https://www.listbox.com/member/archive/rss/303/10443978-6f4c28ac> | > Modify > <https://www.listbox.com/member/?&;> > Your Subscription <http://www.listbox.com> > -- Full employment can be had with the stoke of a pen. Simply institute a six hour workday. That will easily create enough new jobs to bring back full employment. ------------------------------------------- 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
