Information Technology Reference
In-Depth Information
simple reason that cracking such puzzles is precisely the kind of thing humans expect
computers to be able to do, while humans, save for a select few trained in formal
logic, have quite a bit of trouble with such puzzles.
Despite the fact that we anchor our r&d to “trickery,” we do make one assumption
about the internals of a creative T machine, and one general assumption about the
epistemic context of any computational artifact that we produce.
Regarding internals, we assume that the computing machine that is a candidate for
creativity T have at least the minimum representation-and-reasoning power of quan-
tified epistemic logic. This minimality condition, and the methodology that accom-
panies it, have been defended elsewhere [ 5 ]. The condition reflects Bringsjord's
affirmation of Piaget's seminal position that mature, general-purpose human cogni-
tion (including, most prominently, problem-solving) consists of processes operating
largely on formulas having at least the expressivity of formulae in full first-order logic
[ 17 ]. Given the affirmation of the condition in question, the present work reflects a
desire to engineer machines that are creative T + : they are both creative T and their
internal processing conforms to Piagetian concepts of general intelligence and cre-
ative problem-solving. For more on machines that embody the first condition, see
[ 10 ].
What is our second assumption? We assume that computational creativity cannot
be formalized and engineered without yet another nod in the direction of logic, and
specifically epistemic logic. To see our second assumption, consider a person/agent
A who has produced an artifact a through some creative process p . Any formal
model of A 's cognition before, during, and after the creative process should by our
lights have sufficient representational capacity to let us conclude, or at least consider,
whether:
1.
A knows that A itself has produced a inacreativemanner.
2.
A believes that a has not been produced by any other agents.
3.
Aknew what existed and what did not exist before it started the process that produced a .
4.
A desired to create an artifact to fulfill some need.
5.
A intended to engage in a series of actions making up the process p .
The above (informal) micro-theory is admittedly incomplete, but does seem
unavoidable, and suggests that higher cognition surrounding creativity can be suc-
cinctly captured with an expressive formal logic. We do not claim here that we have
invented and specified this logic, but we have taken appreciable steps toward doing
so, by way of our modeling framework,
CEC
, upon which
M
is based. We turn now
to
CEC
and
M
.
14.3 The Music Calculus
While considerable work has been devoted to modeling music at various levels, from
the raw signal-processing stage to representing hierarchical structures, modeling of
the cognitive, social, and doxastic dimensions of music has not been carried out.
Search WWH ::




Custom Search