We used a recurrent neural network as a fitness function for a genetic algorithm to generate monophonic solos.
We used a recurrent neural network as a fitness function for a genetic algorithm to generate monophonic solos. The genetic algorithm is based on GenJam as described in Biles (1994). We conducted training sessions with human participants in order to compare and quantify some of the differences between human-feedback and RNN fitness functions. We found that the RNNs can effectively play the role of human fitness feedback, but still suffer in many areas. Our results suggest that certain types of recurrent neural networks can address the issues with human feedback, and thus should be explored in future research.
Mitrano, P., Lockman, A., Honicker, J., & Barton, S. (2017). Using Recurrent Neural Networks to Judge Fitness in Musical Genetic Algorithms. Proceedings of the 5th International Workshop on Musical Metacreation (MUME) at the 8th International Conference on Computational Creativity (ICCC). https://digitalcommons.wpi.edu/humanitiesarts-pubs/4
*denotes a WPI undergraduate student author