Artificial Intelligence (AI) and Machine Learning (ML) methods are becoming increasingly important in both science and society. In climate science - where complex biophysical and societal processes interact across diverse temporal and spatial scales, and datasets are often large, heterogenous and incomplete - AI and ML methods offer new powerful solutions.
In this July session, Ibrahim Tahri will present insights into AI applications at IIASA and share his professional experiences in the field.
Kai Kornhuber will introduce the seminar and moderate the session.
For online participation, a registration is necessary.
Neural Network Solvers for High-Dimensional PDEs: The Deep Galerkin Method
The Deep Galerkin Method (DGM) presents a novel framework for solving high-dimensional partial differential equations (PDEs) by integrating deep learning with mathematical physics. This mesh-free approach employs deep feedforward neural networks to approximate solutions, reformulating the PDE as an optimization problem that minimizes the governing operator's residual using automatic differentiation and stochastic gradient descent. The architecture effectively captures spatio-temporal dependencies through continuous approximations without specialized recurrent layers. Demonstrated in mathematical finance, DGM mitigates the curse of dimensionality inherent in classical methods and offers broad applicability across computational physics, stochastic control, and engineering. It thus provides a scalable, mesh-free alternative at the intersection of machine learning and scientific computing.
The monthly seminar AI for Climate Science at IIASA will feature global experts in the field of AI and ML who will showcase the newest methodological advancements and applications in the field. Through a series of invited talks, the seminar showcases cutting edge research with the aim of strengthening AI and ML expertise at IIASA and to foster external collaborations. Additionally, it serves as an institute-wide platform for discussions and knowledge exchange across programs and working groups to spark new ideas and innovations.
As an initiative from the ECE/ ICI Theme on Extreme Weather and Climate Dynamics, this seminar is designed for both experts already integrating AI and ML into their workflows and those eager to expand their knowledge in these fields.
Upcoming Events
Hybrid: online and at the Austrian Academy of Sciences
Public lecture: Digitalization and AI within planetary boundaries
Chulalongkorn University, Thailand