On 9/5/06, M. Riad <[EMAIL PROTECTED]> wrote:
>
> Sorry to barge into the conversation in this way, but YKY mentioned something I needed clarification with.
>
> You said:
>
> With logic I can write down a rule for recognizing this pretty easily, mainly due to the use of symbolic variables. So you see the compressive power of logic.
>
> But wouldn't this mean you'd end up "hand-coding" a lot of logic rules? I don't really follow how this could achieve generic AI. Am I missing something?
>
> Sorry to barge into the conversation in this way, but YKY mentioned something I needed clarification with.
>
> You said:
>
> With logic I can write down a rule for recognizing this pretty easily, mainly due to the use of symbolic variables. So you see the compressive power of logic.
>
> But wouldn't this mean you'd end up "hand-coding" a lot of logic rules? I don't really follow how this could achieve generic AI. Am I missing something?
Learning is actually the most difficult / most open-ended aspect in AGI. Most people equate machine learning with things like neural networks or support vector machines, but logic can be a substrate for learning too. One example is inductive logic programming (ILP) which can learn Prolog-like rules from examples. For example, we can present 10 examples of "bottle" to the AGI and it will learn a logical rule that defines bottles
in general. The logical kind of machine learning is actually more similar to human learning because it requires relatively fewer examples (unlike NN training).
The Pattern Recognizer itself only applies logic rules for pattern recognition. The "Inductive Learner", which is on the lower-left of my arhitecture diagram, would be the module responsible for learning the definition of those patterns. My architecture separates the Learner from other modules.
Another way is to insert the rules by hand, as a last resort. This is actually the route taken by Cyc. So we can import Cyc's knowledgebase too.
YKY
To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/[EMAIL PROTECTED]
