Maimai Chart Generation Model Based on Machine Learning Algorithms

Abstract

Chart generation for rhythm games is a demanding task that requires specialized design knowledge. In the past, there were several models proposed to generate charts by AI for different rhythm games. However, no full model has yet been designed for Maimai DX. In this paper, the author presents a model with high modularity. The main research approach is based on prior models and fits existing algorithms into the model presented by the author. The main purpose of this paper is to propose a possible chart generation model for Maimai DX and to propose improvements for future iterations. The model is designed by blocks, and each block is composed of specific goals and algorithms. The design is characterized by a high degree of modularity in the model, which provides space for future iterations because parts can be easily replaced.