【114/9/12】11:00-12:00 王士欣 博士 (PhD Candidate in Mathematics, University of Utah, USA)
發佈日期 :
2025-09-08
| Seminar | |
|
|
|
| Date | 2025-9-12 11:00-12:00 |
| Place | 數學館3樓會議室 |
| Speaker | 王士欣 博士(PhD Candidate in Mathematics, University of Utah, USA) |
| Title | What Can Mathematicians Do in Deep Learning? A Journey from Pure Math to AI |
| Abstract | In this talk, I will share my journey from algebraic geometry to machine learning, and how a foundation in pure mathematics has guided me into areas such as geometric deep learning and flow matching. I will discuss how a mathematical perspective enables us to recognize fundamental problems, reformulate them rigorously, and develop solutions with strong theoretical foundations. Beyond my own path, I will reflect on the broader role mathematicians can play in the era of deep learning—clarifying the foundations of modern AI, ensuring theoretical soundness, and inspiring novel models and methodologies that push the field forward. |
|
|
|