Title: Threshold Designer Adaptation: Improved Adaptation for Designers in Co-creative Systems

Abstract: To best assist human designers with different styles, Machine Learning (ML) systems need to be able to adapt to them. However, there has been relatively little prior work on how and when to best adapt an ML system to a co-designer. In this paper we present threshold designer adaptation: a novel method for adapting a creative ML model to an individual designer. We evaluate our approach with a human subject study using a co-creative rhythm game design tool. We find that designers prefer our proposed method and produce higher quality content in comparison to an existing baseline.