Speaker: Dr. Nicole Yang
From: Emory University
Abstract
Dynamical systems are widely used to model complex real-world phenomena. They are models based on physical principles, which are highly interpretable but have limits in expressive capabilities. Furthermore, the already high dimensional, and complicated behavior make them difficult to analyze and solve. In recent years, data-driven models have excelled at handling complex, high-dimensional data. However, these models face challenges in their interpretability, reliability, robustness, and computational complexity. My vision is to connect stochastic dynamical systems and statistical learning to bridge the gap between physics-based models and data-based models.
In this talk, I present some of my work demonstrating the potential in addressing dynamical systems and statistical learning problems wholistically. I first focus on understanding and learning dynamical systems with stochastic, interactive, and chaotic behavior. I present my work on decision-making in complex systems through mean-field games and stochastic control. Next, I discuss learning guided by dynamical systems. In particular, these approaches tackle challenges in generative models and enable flexible representations of transportation between complex distributions.
For more info, please follow this link.
Read More