January 27, 2025 12:00 PM / Students
FIGSS Seminar: Unveiling Epidemic Spread Dynamics with Physics-Informed Deep Learning
Interdisciplinary seminar for PhD students and students at FIAS

Venue: FIAS Faculty Club
The interdisciplinary FIGSS seminar takes place every second Monday during the semester from 12:00-13:00. PhD students of the Graduate School are given the opportunity to present and discuss their research results here. External speakers are also invited from time to time.
Program on 27.01.2025
Speaker: Shuai Han; Zhou Group
Abstract:
Studying the dynamics of epidemic spread is crucial for disease control and public health decision-making.
Traditional epidemiological models rely on predefined parameters, making it challenging to accurately capture the spatiotemporal variations of an outbreak.
Meanwhile, deep learning approaches, while powerful in data-driven pattern recognition, often lack interpretability and fail to reflect the underlying transmission mechanisms.
Integrating physics-based principles with deep learning addresses these limitations, enabling models to learn complex propagation patterns while preserving interpretability.
Our research focuses on bridging epidemiological modeling and deep learning to explore the intricate dynamics of epidemic spread, gradually unveiling the hidden mechanisms behind epidemic propagation.
By combining epidemiological insights with data-driven modeling, physics-informed deep learning effectively captures spatiotemporal transmission characteristics, enhances forecasting accuracy, and provides valuable guidance for epidemic control and intervention strategies.