I have no doubt that *eventually* they will get to that. (Numenta's technology is one approach to that -- not that it necessarily works, since I haven't personally used it, but failures are usually encountered on the road to success.) But because of the potentially long road ahead for such research, and likely much wasted time dealing with what I would call the "operating system" of the brain as opposed to the important parts of the "program", I find it more probable that the top-down approach to reverse engineering the mind will beat the bottom-up approach to the punch.
On Aug 29, 2012 8:39 AM, Mike Tintner <[email protected]> wrote: Surely, the key thing is we still don’t understand the ‘engram’ – how information is laid down in neurons (or elsewhere).- the very foundation of how the brain works. Without that, it’s hard to see any neuron-based approach to AGI as other than mad, no? In general, whenever s.o. talks of a “science-based” approach to AGI, it’s b.s. From: [email protected] Sent: Wednesday, August 29, 2012 1:56 PM To: AGI Subject: Re: [agi] LM741 You said: I don't see how mastery of the neuron will help anyone towards AGI. It's called reverse engineering. You have to start somewhere. It seems obvious that the best places to start are with what are most readily visible: either the top-level behavior, or the bottom-level components. I prefer starting from the top, as do you apparently, but there's no reason to assume they aren't working on exactly the same problem as us, regardless of what you think of their egos. (Your argument that they must have huge egos applies equally as well to everyone else on this list, including you and me.) Numenta is a good example of how understanding neurons could lead to AGI (and is therefore a relevant subject for AGI-related lists like this one), even though they've got a long way to go and a lot of improvement to make. On Aug 29, 2012 4:29 AM, Anastasios Tsiolakidis <[email protected]> wrote: On 29/08/2012, at 10.52, Ian W Eberhart <[email protected]> wrote: Anastasios, I don't think you've actually said anything yet, A blessing in disguise! What I have said in the general sense, on and off the list, is that I would like to see a tight AGI focus, ie if the answer to a question is not seen as critical to AGI then we should keep it for another forum. Specifically I don't see how mastery of the neuron will help anyone towards AGI. The same could be said about LM741' unless someone is seriously considering stringing together a few billion of them so it seems a little rough to accuse someone of egotism when all they are trying to do Let's be clear about one thing, it takes a huge ego to work with anything intelligence related. I scolded the fuck and other miscelania. And I do have a pet peeve towards neurodudes and dudettes who tell me how amazingly many things they know and prove, they can share their glories with all the worlds autistic, dyslexic, blind, deaf, bipolar, Alzheimer's etc sufferers. AT 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-c97d2393 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-2484a968 Powered by Listbox: http://www.listbox.com
