If logic (traditional logic) is used to refer to numerical values (like
weighted values) then the evaluation of various logical combinations will
require some numerical system that would be (or might be) governed by logic but
which exists outside the logic representing the values. Similarly, if a logical
system refers to things and events, as relationships between those things and
events are noted or observed they have to be characterized by logical
subsystems. The methodology of how those concepts are represented within the
system has to be characterized by a systematic approach that is external to the
actual traditional logic that would then be used to characterize those events.
The belief that the limitations of simple traditional logic can be used to
prove that an approach to AI/AGI is therefore narrow by definition would have
to be based on a highly simplified idea about traditional logic. Those kinds of
constraints were never a property of applied logic and AGI has to be able to
refer to something outside its initial representational values. Many ideas (or
at least a lot of ideas) about computer science were designed with closed
algorithms in mind. The majority of numerical computations are designed with
closed algorithms. To immediately say that any constraint of a closed numerical
algorithm defines the limitation of AI programs would not make any sense
because we are not talking exclusively about closed numerical algorithms. We
may talk about them but that is not what we are generally talking about.
------------------------------------------
Artificial General Intelligence List: AGI
Permalink:
https://agi.topicbox.com/groups/agi/T1675e049c274c867-M278506d61bd5c3bd3d9b89ec
Delivery options: https://agi.topicbox.com/groups/agi/subscription
