Overview

Partial Differential Equations (PDEs) are fundamental to modelling a wide range of physical, biological, and engineered systems. With recent advances in machine learning, researchers have begun to explore how data-driven methods can learn and approximate PDEs directly from observations, offering new tools for scientific discovery.

This meetup will highlight recent developments in the learned PDEs using AI techniques. We will discuss foundational methods such as neural operators and physics-informed neural networks (PINNs), explore applications across disciplines, and examine the open challenges in interpretability, generalization, and scalability. The session aims to provide attendees with both a theoretical understanding and practical insights into this emerging area.

After attending this meetup, participants will be able to:
● understand how machine learning can be used to model and solve PDEs
● gain an overview of key methods like PINNs and neural operators
● explore real-world applications in fluid dynamics, materials science, and more
● identify challenges and research opportunities in learned PDEs

Angelica I Aviles-Rivero

Tsinghua University

Bin Dong

Peking University

Lu Lu

Yale University

Co-Organising Committee

Alphabetic Order

Chun-Wun Cheng

University of Cambridge

Yanqi Cheng

University of Cambridge

Chao Deng

YMSC, Tsinghua University

Haonan Dong

Tsinghua University

Yanming Xia

Tsinghua University

Xuxiang Zhao

Tsinghua University

Keynotes

Alphabetic Order

Hongyu Liu

City University of Hong Kong

Lu Lu

Yale University

Youjia Wu

Bosch Center for Artificial Intelligence

Qi Meng

Chinese Academy of Sciences

Flash Talks

Alphabetic Order

Scientific Communicator

Yanya Cheng

Volunteers

Alphabetic Order

Yaqi Gao

Nan Huang

Songlin Zhou