The training issue is a real one, but presumably over time electronics that would be part of these wetware/hardware combination brains could be developed to train the wetware/hardware machines --- under the control guidance of external systems at the factory --- relatively rapidly, so that in say one year the brains know as much as a bright teenager. Since all this training could take place in parallel, it would not be that great of an overall cost.
I personally would prefer all electronic brains where you could have much more of an ability to examine virtually any part of the system. But it is possible that it will be quite a while before we can develop electronics as cheap and as power efficient as neurons. (Of course it is also possible that electronic advances will happen so fast that there is no real reason for using wetware.) Ed Porter -----Original Message----- From: Mike Archbold [mailto:[EMAIL PROTECTED] Sent: Thursday, August 14, 2008 5:24 PM To: agi@v2.listbox.com Subject: Re: [agi] Meet the world's first robot controlled exclusively by living brain tissue > 2008/8/14 Ciro Aisa <[EMAIL PROTECTED]>: >> On Thu, Aug 14, 2008 at 10:25:57AM +0100, Bob Mottram wrote: >>> I doubt that there will be much practical application of biological >>> neuron powered robots, since the overhead of keeping the biology alive >>> would be too troublesome (requiring feeding and removal of waste >>> products), >> >> Actually, better than spending X billion dollars... > > > You would need train each neural net individually, rather like > educating humans, and this woudn't be very economical compared to a > pure machine approach where the state of one system can be coppied an > arbitrary number of times at high speed with no information loss. > Wouldn't it be just ordinary evolutionary-survival-darwinian... I guess there would have to be some means of rat-brain reproduction. So, you'd need to leave the reproduction functionality in there. Aside from sounding like that Spock's Brain episode, it raises the troubling issue that the resulting organism is is an organism as usual. Still haven't achieved AGI! Suppose you wanted to start making IBM mainframe programmer wetware. Once you've got that DNA worked out, what's the difference between that and a real mainframe programmer? Actually this would be a good first example, since it hasn't changed since about 1975. Mike > > ------------------------------------------- > agi > Archives: https://www.listbox.com/member/archive/303/=now > RSS Feed: https://www.listbox.com/member/archive/rss/303/ > Modify Your Subscription: > https://www.listbox.com/member/?&; > Powered by Listbox: http://www.listbox.com > ------------------------------------------- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?&; Powered by Listbox: http://www.listbox.com ------------------------------------------- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244&id_secret=111637683-c8fa51 Powered by Listbox: http://www.listbox.com
