Xiaoyu (Coco) Huang (Temple)
Event Date
2025-03-26
Event Time
02:30 pm ~ 04:00 pm
Event Location
Wachman Hall 412
Abstract: In this talk, we will discuss some recent applications of machine learning to number theory. In particular, we will discuss murmurations of Arithmetic L-functions and experiments on the Tate–Shafarevich groups. We will also introduce the recent results of applying transformer models to predict information about Frobenius traces a_p from elliptic curves given information about other traces a_q. Our experiments reveal that these models achieve high accuracy, even in the absence of explicit number-theoretic tools like functional equations of L-functions. We also present partial interpretability findings on the patterns learned by the machine learning models. No prior knowledge of machine learning is assumed.