--- Mike Tintner <[EMAIL PROTECTED]> wrote: > AN AGI MACHINE - a truly adaptive, truly learning machine - is one that will > be able to: > > 1) conduct a set of goal-seeking activities > > - where it starts with only a rough, incomplete idea of how to reach its > goals, > > - i.e. knows only some of the steps it must take, & some of the rules that > govern those steps > > - and can find its way to its goals "making it up as it goes along" > > - by finding new ways round more or less unfamiliar obstacles. > > To do this it must be able to: > > 2) Change its steps and rules - > > -not just revising them according to predetermined formulae but > > -adding new steps and rules, & even > > -creating new rules, that break existing ones. > > 3) can learn new related activities
Consider the following machine. At each time step, the machine outputs a 0 or a 1, then receives a 0 or 1 from an unknown environment. The goal of the machine is to predict the next input bit. Which of the following is AGI? 1. The machine guesses randomly. 2. The machine always guesses 0. 3. The machine always guesses whatever the last input was. 4. The machine keeps a record of past inputs. It compares the last n inputs with the history string and predicts whatever input bit followed the most recent match. If no match is found, it tries n-1, n-2, ..., down to 0. 5. Like 4, but includes its own output bits interleaved with the input bits. 6. The machine enumerates all Turing machines by increasing length and runs each one for time limit t until one machine outputs the entire (interleaved) history string plus one more input bit, then predicts that bit. -- Matt Mahoney, [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=4007604&user_secret=8eb45b07
