Modeling Unknown Stochastic Dynamical System via Autoencoder
Speaker: Zhongshu Xu (OSU) Dates: 2024/10/03 Location: MW 154 Abstract: This talk presents a novel data-driven method for modeling unknown stochastic dynamical systems using autoencoders. The approach focuses on learning the flow map of an underlying system by identifying unobserved…
Dynamical System Learning via Neural Networks
Speaker: Zhongshu Xu (OSU) Dates: 2023/02/09 Abstract: This talk will briefly review Prof Dongbin Xiu’s work on dynamical system learning and Neural Networks. First, I will discuss the connection between Euler forward scheme and ResNet, and why multi-step ResNet can…
Butterfly Factorization and Butterfly-Net: from Numerical Linear Algebra to Machine Learning
Speaker: Zhongshu Xu (OSU) Dates: 2022/10/06 Abstract: Butterfly Factorization is a data-sparse nearly optimal approximation for the discrete Fourier Integral Operator (FIO) matrices. It is constructed based on interpolative low-rank approximations of the complementary low-rank matrix. For an N ×N…