I quickly read the article (in 1/2 an hour) but got most of the gist of it. I want to read it again more carefully.
The "Core AGI Hypothesis" sounds correct. The desired end result: some approximation of something like human-level intelligence seems to be the goal of all approaches. But after that the problems start. Actually doing this is the problem! :) I think an implied or stated intent of the paper is to find some fundamental processing/solution core that could be shown to be objectively legitimate, and must be included within any solution, regardless of which heading (symbolic, emergent, hybrid) if falls under. The approaches to AGI vary wildly. How can one know which is correct, if there is no AGI really working? But there must be some core assumptions that each approach shares, even when comparing something as drastically different as CYC and AIXI. Mike A On 6/26/14, Anastasios Tsiolakidis via AGI <[email protected]> wrote: > G-d, a lot of broken English in that article, let me guess, 13 yo Ukrainian > or PLN? Certainly the first paragraph on page 2 cannot end in "that > display". > AT > > ------------------------------------------- > AGI > Archives: https://www.listbox.com/member/archive/303/=now > RSS Feed: https://www.listbox.com/member/archive/rss/303/11943661-d9279dae > 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/303/21088071-f452e424 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657 Powered by Listbox: http://www.listbox.com
