At this point I'd describe intelligent machine as being able to construct incremental internal models of wide variety of processes, based on moderately explicit information about those processes (in other words, being able to learn sufficiently general models, and to elaborate them). Using these models, it's able to represent information about other, compound processes, in sufficiently explicit form, and so on. For example, at first it can learn about letters, then words, then about reading, then pick a textbook on some subject, etc. Almost all of functionality of intelligent machine consists in ability to translate data in variety of formats between representations, making various aspects of it explicit. As a result, it can translate consequences of its actions into changes in external processes, and intelligently choose these actions.
-- Vladimir Nesov mailto:[EMAIL PROTECTED] ----- This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=8660244&id_secret=85157164-7cef7f
