ConBreO: a music performance rendering system using hybrid approach of IEC and automated evolution

Authors:
Makoto Tanji;Hitoshi Iba
Affiliations:
University of Tokyo, Tokyo, Japan;University of Tokyo, Tokyo, Japan
Venue:
Proceedings of the 12th annual conference on Genetic and evolutionary computation
Year:
2010

Citing 10
Cited 0

Genetic programming: on the programming of computers by means of natural selection

Genetic programming: on the programming of computers by means of natural selection
Evolving robot behavior via interactive evolutionary computation: from real-world to simulation

Proceedings of the 2001 ACM symposium on Applied computing
A Taxonomy of Global Optimization Methods Based on Response Surfaces

Journal of Global Optimization
Information Theory, Inference & Learning Algorithms

Information Theory, Inference & Learning Algorithms
Pattern Recognition and Machine Learning (Information Science and Statistics)

Pattern Recognition and Machine Learning (Information Science and Statistics)
EFIT-V -: interactive evolutionary strategy for the construction of photo-realistic facial composites

Proceedings of the 10th annual conference on Genetic and evolutionary computation
Eye on the prize: using overt visual attention to drive fitness for interactive evolutionary computation

Proceedings of the 10th annual conference on Genetic and evolutionary computation
Classification and regression-based surrogate model-assisted interactive genetic algorithm with individual's fuzzy fitness

Proceedings of the 11th Annual conference on Genetic and evolutionary computation
Capturing aesthetic intention during interactive evolution

Computer-Aided Design
Interactive Evolutionary Computation-Based Hearing Aid Fitting

IEEE Transactions on Evolutionary Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents an IEC (Interactive Evolutionary Computation) system named ConBreO to render expressive music performance using Genetic Programming. The central problem of IEC is the limitation of number of fitness evaluations because of user fatigue. In the system, we introduce two support techniques for IEC. The first one is a hybrid approach of IEC and automated evolution which allows the system to evolve both of IEC and automated evolution. The second one is the selective presentation which selects a new individual to be evaluated by the user based on its expected improvement of fitness. Using the system, obtained expression rule won an award at a performance rendering contest which evaluates computer systems generating expressive musical performances. Our experiment shows that the selective presentation reduces the number of fitness evaluations required to construct the fitness prediction model and prevents the system evaluating unfruitful individuals.