You do not need and you probably could not determine how digital programming logic works only by emulating the underlying physics of a computer. Even if you could understand how the underlying electronics worked, without the easy availability and utilization of programming languages few of you (none of you who will read this in the next few days) would get very far programming a computer based on a knowledge of the electronics alone. Think how long it would take you to figure out what each instruction does, once you figured out how to provoke an instruction, if you did not have a programming compiler or interpreter to see what was happening. The thought that you are going to create a science based on brain emulation is not realistic. I don't know what will happen thirty or fifty or a hundred years from now but the first step is to make AI programs work in ways that are more thoughtful.
Jim Bromer On Fri, Jun 26, 2015 at 4:50 PM, Dorian Aur <[email protected]> wrote: > " The computational structures of the brain cannot be efficiently > emulated using conventional computing substrates. The implementations > reflect the structure and capabilities of the fundamental materials they > are built from. A scanned brain AGI would be obscenely inefficient on > silicon...." > > Indeed, the above point is highly important, that's why an intermediate > step is required - the hybrid system -http://arxiv.org/abs/1411.5224 and > the entire construct of IGI starting with hybrid systems design makes sense > > Dorian > > On Fri, Jun 26, 2015 at 1:36 PM, J. Andrew Rogers <[email protected]> > wrote: > >> >> > On Jun 26, 2015, at 12:43 PM, Tim Tyler <[email protected]> wrote: >> > >> > On 26/06/2015 14:01, Dean Pomerleau wrote: >> >> The reason I think this interesting and relevant is that many high >> profile people (e.g. singularity economist Robin Hanson) see WBE as the >> most likely path to AGI because on the surface it seems like all that is >> required is straightforward engineering - all we need is better scanning >> methods (e.g. extending vitrification techniques already apparently quite >> good for mice to work for human-sized brains), straightforward extension of >> current neural modeling techniques (e.g. refinement to Hodgkin-Huxley model >> of neurons) and more powerful computers on which to run the emulations. >> > >> > I never bought into this. Areoplanes aren't scanned birds. The motor >> car wasn't >> > a scanned horse and cart. Submarines aren't scanned fish. Computers >> aren't >> > scanned brains. >> >> >> This is such an important point. >> >> Even if we had the necessary brain scanning technology, the computational >> structures of the brain cannot be efficiently emulated using conventional >> computing substrates. The implementations reflect the structure and >> capabilities of the fundamental materials they are built from. A scanned >> brain AGI would be obscenely inefficient on silicon to the point where it >> might not even have economic value (e.g. within the physics and engineering >> limits of silicon fabrication the emulation could be much slower than an >> actual human brain). >> >> There is a reason no one does their computing by emulating a Turing >> machine on top of conventional silicon. While the computing models are >> theoretically equivalent, there is an extreme performance penalty for >> moving too far from the physical model of computation, even if the model is >> amenable to an optimized silicon implementation. >> >> >> >> >> ------------------------------------------- >> AGI >> Archives: https://www.listbox.com/member/archive/303/=now >> RSS Feed: >> https://www.listbox.com/member/archive/rss/303/17795807-366cfa2a >> 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> > <https://www.listbox.com/member/archive/rss/303/24379807-653794b5> | > Modify > <https://www.listbox.com/member/?&;> > Your Subscription <http://www.listbox.com> > ------------------------------------------- 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
