Research 

Physics

​

I am interested in X-ray spectroscopy and theoretical modeling.

In 2023, I studied reflection spectra with uneven vertical density, though the work was not published in the end.

In 2024, I became fascinated by machine learning and began exploring its applications in solving physical problems. I experimented with CNNs, RNNs, and Transformer-based models, and incorporated attention mechanisms for classifying galaxy observation data. Toward the end of 2024, I used StyleGAN3 to generate redshift images of black holes and also developed a neural-network–based version of the xillver model.

 

That same year, I published my first paper, X-ray reflection spectroscopy with improved calculations of the emission angle.

​

In 2025, I continue to pursue research at the intersection of machine learning and accretion disk physics. I am currently developing a new public reflection model, DAO, designed to be community-friendly and broadly beneficial to the field. There remain many exciting directions to explore with this model. This year, I am also exploring the potential of reinforcement learning in astrophysics, which may offer efficient ways to handle large volumes of data.