Generative Melody Composition with Human-in-the-Loop Bayesian Optimization
This addresses the problem for novice composers using generative models, but it is incremental as it applies existing optimization techniques to a specific domain.
The paper tackles the challenge of finding desired melodies in the high-dimensional latent space of deep generative models by introducing an interactive system that uses human-in-the-loop Bayesian optimization, where the system generates candidates and users provide preferential feedback iteratively, with a pilot study suggesting its potential.
Deep generative models allow even novice composers to generate various melodies by sampling latent vectors. However, finding the desired melody is challenging since the latent space is unintuitive and high-dimensional. In this work, we present an interactive system that supports generative melody composition with human-in-the-loop Bayesian optimization (BO). This system takes a mixed-initiative approach; the system generates candidate melodies to evaluate, and the user evaluates them and provides preferential feedback (i.e., picking the best melody among the candidates) to the system. This process is iteratively performed based on BO techniques until the user finds the desired melody. We conducted a pilot study using our prototype system, suggesting the potential of this approach.