Geometric Deep Learning
Автор: Society for Industrial and Applied Mathematics (SIAM)
Загружено: 2024-03-21
Просмотров: 1554
Geometric Deep Learning is a general blueprint unifying different neural network architectures through the fundamental principles of symmetry and invariance. In this tutorial, we will focus on a particular instance of Geometric Deep Learning: Graph Neural Networks (GNNs). We will describe the anatomy of GNNs and motivate different architectural choices. In the second part of the tutorial, we will relate GNNs to differential equations on graphs. This will allow us to look at GNNs through the lens of differential geometry and algebraic topology.
MT5: Geometric Deep Learning
Organizer: Michael Bronstein
University of Oxford, United Kingdom
This talk was given at the 2022 SIAM Conference on Mathematics of Data Science in San Diego, California, U.S. Learn more about SIAM Conferences at https://www.siam.org/conferences/abou...
Доступные форматы для скачивания:
Скачать видео mp4
-
Информация по загрузке:
