A Data Scientific Approach to Elucidate Multi-scale Physics of Galaxies in the Cosmos

在过去的一年里，我的研究集中在三个主要领域：星系流形和恒星形成历史估计。利用无监督机器学习，我成功地确定了两个基本参数，它们准确地代表了星系，与目前的理解一致。此外，我开发了一种使用生成模型来模拟恒星形成历史的方法，使之与观测更好地匹配，并促进了我们对星系形成历史的理解。此外，我在宇宙磁性项目上与SKA-日本科学工作组合作，通过模拟探索银河系磁性的可观测性，并研究解开复杂磁场结构的替代方法。

Over the past year, my research has focused on three primary areas: the Galaxy Manifold and star formation history estimation. Utilizing unsupervised machine learning, I successfully identified two fundamental parameters that accurately represent galaxies, aligning with current understanding. Furthermore, I developed a method employing generative models to simulate star formation histories, enabling a better match with observations and advancing our comprehension of galaxy formation history. Additionally, I collaborated with the SKA-Japan Science working group on the cosmic magnetism project, exploring the observability of galactic magnetism through simulations and investigating alternative approaches to disentangle complex magnetic field structures.