Akuiro
周朕葵

I am an undergraduate student in the School of Physics at Peking University in Beijing, China.

I am currently working on an undergraduate research project on machine-learning-assisted screening of Hall responses in topological quantum materials, under the guidance of Prof. Ji Feng.

Email  /  CV  /  GitHub  /  Phone

Akuiro

Education 教育经历

Peking University, School of Physics Sep. 2024 - Jun. 2028 (Expected)

B.S. in Physics

  • Selected coursework: Theoretical Mechanics, Modern Physics, Optics, Mathematical Methods in Physics I & II, Probability and Statistics, Electromagnetism, and Data Structures and Algorithms.

Research 科研经历

My current interests include topological quantum materials, Hall transport, first-principles calculations, and machine learning for materials research. My undergraduate project is currently in the literature-review and methodology-preparation stage.

Machine-Learning-Assisted Screening of Hall Responses in Topological Quantum Materials Apr. 2026 - Present

Independent Undergraduate Research Project · Advisor: Prof. Ji Feng, Peking University

  • Investigating how materials databases and machine-learning methods can help screen candidate topological quantum materials for quantum anomalous Hall and thermal Hall responses.
  • Reviewing topological band theory, Berry curvature, and Hall transport; surveying available materials databases; and preparing methods for DFT data extraction and materials-descriptor construction.

Projects 项目经历
QPDS interface QPDS: Quantitative Poker Decision System
Python, Flask
  • Built a modular decision system with opponent-range parsing, Monte Carlo equity simulation, and exact enumeration for small state spaces.
  • Combines equity, pot odds, effective SPR, position, board texture, and opponent behavior to return action recommendations, expected value, and confidence through a Flask REST API.
Modern Art digital assistant interface Modern Art: Board Game Digital Assistant
Python, Streamlit
  • Implemented auction-state management for 3-5 players across four rounds, including fifth-painting termination, cumulative artist values, round settlement, and final asset rankings.
  • Added a bilingual interface, transaction logs, JSON save/load, undo support, and artist-value history charts.

Personal Blog Deployment and Migration

Docker Compose, Nginx, GitHub Actions, Linux

  • Deployed MX-Space, MongoDB, and Redis with Docker Compose; configured Nginx reverse proxying; and automated the Shiro frontend build and SSH/PM2 release workflow with GitHub Actions.
  • Migrated the personal homepage, articles, and project records to GitHub Pages after the previous server and domain expired.

Notes 随笔
  • 偏微分方程计算方法 2026-07-22 · 核心主题：有限差分法、基矢展开法（含 DVR）、B-样条、小波分析、有限元法、复杂边界条件处理
  • 如何优雅地部署自己的个人博客主页 🚀 2026-07-22 · 2025年7月15号，作为一个爱水群的入，被群里一位学长炫酷的个人主页所吸引（询问后发现是 Shiro 主题 ）于是开启一段小白搭配AI配置…
  • 微分方程计算方法 2026-07-22 · 💡 物理直觉：初值问题是"知道现在，推演未来"（动力学），边值问题是"知道两端，反推中间"（稳态分布）。
  • 数值微分与积分 2026-07-22 · 用离散点的函数值近似连续的微分/积分，通过多项式插值作为桥梁——先从离散点构造插值多项式，再对多项式求导/积分获得数值微分/积分公式。
  • 本征值问题 2026-07-22 · 本质上 主线就是 把物理问题 变成 矩阵的本征值问题，然后利用正交变换把矩阵“变得简单但保证其本征值不变”，再利用迭代法吧本征值计算出来！
  • 神经网络与机器学习 2026-07-22 · 机器学习已发展成为一种新的科学范式——当物理方程难以描述或求解时，用数据驱动的方法进行推断。物理学中各领域 ML 论文数量爆发式增长：
  • 蒙特卡洛方法 2026-07-22 · 通过统计采样（随机抽样）解决传统方法难以处理的数学/物理问题——积分、优化、多自由度系统模拟等。

Honors & Awards 荣誉与奖项
  • Academic Excellence Award, Peking University · 2024-2025 Academic Year
  • School of Physics Merit Award, Chinese Undergraduate Physics Tournament (CUPT) · Jun. 2025
  • Silver Medal, Chinese Physics Olympiad Final (CPhO) · Oct. 2023