Joint Math-Stat Colloquium: Itai Dattner (Haifa)
Location
Mathematics/Psychology : 103
Date & Time
September 27, 2024, 11:00 am – 12:00 pm
Description
Title: Advances in Scientific Machine Learning: Theory, Methods, and Applications
Abstract: In this lecture, we explore the cutting-edge advancements in scientific machine learning, focusing on the integration of theoretical foundations, methodological innovations, and practical applications. We delve into the core concepts of parameter estimation, model selection, and statistical testing of differential equation models under conditions of noisy and sparse data. Our discussion will encompass minimax adaptive nonparametric statistical testing, parameter estimation through smoothing techniques, identifiability, consistency, and rates of convergence.
We will illustrate these theoretical principles through diverse applications: enhancing the quality assessment of agricultural produce using physics-guided neural networks, modeling and estimating motion synchronization in psychotherapy and communication disorders, hybrid-AI quantile estimation and forecasting in climate and infectious diseases, optimizing hemodialysis treatment, and developing a digital twin for the Sea of Galilee. This multidisciplinary approach highlights the transformative potential of scientific machine learning across various domains, emphasizing its role in addressing complex real-world challenges.
We will have the Departmental Coffee and Tea from 10 to 10:45 in M&P 422.