Artificial Creativity: Improving on Algorithmic Music Composition Using Genetic Algorithms

A program using rules of music theory was written to create original compositions. Following the rules of music theory guarantees harmonious compositions, but certain aspects of musical composition cannot be defined by music theory. It is in these aspects of musical composition where the human mind uses creativity. The current research utilizes a genetic algorithm to make decisions that cannot be made using music theory alone. To test the effectiveness of the approach, subjects were interviewed about certain musical preferences, and these preferences were used as a basis for the fitness evaluation function of the genetic algorithm. Subjects listened to an initial composition written by the program prior to learning and then one composed by the program after learning a style based on the subject’s preferences. Each composition was ranked by the subjects on a scale from 1 to 10. Overall, subjects preferred the learned compositions over the initial compositions. Key-Words: Machine learning, genetic algorithms, artificial creativity, artificial intelligence, music