Abstract
Victor Frankenstein sought to create an intelligent
being imbued with the rules of civilized human conduct,
who could further learn how to behave and possibly even
evolve through successive generations into a more
perfect form. Modern human composers similarly strive
to create intelligent algorithmic music composition
systems that can follow prespecified rules, learn
appropriate patterns from a collection of melodies, or
evolve to produce output more perfectly matched to some
aesthetic criteria. Here we review recent efforts aimed
at each of these three types of algorithmic
composition. We focus particularly on evolutionary
methods, and indicate how monstrous many of the results
have been. We present a new method that uses
coevolution to create linked artificial music critics
and music composers, and describe how this method can
attach the separate parts of rules, learning, and
evolution together into one coherent body.
Users
Please
log in to take part in the discussion (add own reviews or comments).