Elucidation of the adaptive mechanism of intricate human motion imitated by deep reinforcement learning

点击翻译按钮获取中文摘要

In the first year of our project, we focused on analyzing dance motion data using our newly developed low-dimensional representation algorithm. Our goal was to gain a better understanding of the adaptive motor skills, coordination patterns, motion consistency, and complexity of various dance genres, by examining them from both kinematic and kinetic perspectives. We were able to create computational models that can accurately reproduce dance motor skills through deep reinforcement learning, which was a significant achievement. Our approach allowed us to explore the complex interplay between the cognitive and physical aspects of dance, and to gain valuable insights into how dancers develop and refine their skills over time.