AIMS Home

Data-driven methods in dynamical systems

Organizer(s):
Name:	Affiliation:	Country:
Ruhui Jin	University of Wisconsin-Madison	USA
Shi Chen	University of Wisconsin-Madison	USA
Qin Li	University of Wisconsin-Madison	USA

Abstract:

Dynamical systems arise in a broad range of interdisciplinary fields, describing the evolutions of climate, networks, neuroscience and physical models. With advanced data-driven methodology, there has been significant development in learning the underlying governing equations from data. Theoretical and computational challenges accordingly arise. This session aims to present recent works in data-driven dynamical system learning and the related areas including machine learning, inverse problem and uncertainty quantification.

List of approved abstract
Name	Affiliation	Title
Ke Chen	University of Maryland at College Park	Deep Operator Learning Lessens the Curse of Dimensionality for PDEs
Yifan Chen	Caltech	Gradient flows for sampling: affine invariance and numerical approximations
Shi Chen	University of Wisconsin-Madison	Global Convergence of Gradient Descent for Multi-Layer ResNets with Homogeneous Activation Functions in the Mean-Field Regime
Victor Churchill	The Ohio State University	Learning the Evolution of Unknown Systems via Deep Neural Networks
Shukai Du	University of Wisconsin-Madison	Fast, low-memory methods for radiative transfer via hp-adaptive mesh refinement
Longxiu Huang	Michigan State University	Bandlimited Graph Signal Recovery from Randomized Space-time Samples
Ruhui Jin	UW-Madison	Tensor-structured sketching for constrained optimization
Tulin Kaman	University of Arkansas	Advancements in reduced order modeling and physics-informed neural networks for solving large scale partial differential equations
Charles Kulick	University of California, Santa Barbara	Scalable Multi-Species Agent-Based Modeling with Sparse GP
Changhong Mou	University of Wisconsin-Madison	Combining Stochastic Parameterized Reduced-Order Models with Machine Learning for Data Assimilation and Uncertainty Quantification with Partial Observations
Yinling Zhang	University of Wisconsin Madison	A Causality-Based Learning Approach for Discovering the Underlying Dynamics of Complex Systems from Partial Observations with Stochastic Parameterization