webinar register page

IMA Data Science Seminar - Nadav Dym
Speaker: Nadav Dym
Title: Efficient Invariant Embeddings for Universal Equivariant Learning
Abstract: In many machine learning tasks, the goal is to learn an unknown function which has some known group symmetries. Equivariant machine learning algorithms exploit this by devising architectures (=function spaces) which have these symmetries by construction. Examples include convolutional neural networks which respect the translation symmetry of images, and neural networks for graphs or sets which respect their permutation symmetries. More examples will be discussed.

Jan 31, 2023 01:25 PM in Central Time (US and Canada)

* Required information
Loading

Speakers

Nadav Dym
Assistant Professor @Technion-Israel Institute of Technology
Nadav Dym is a senior lecturer (assistant professor) in the Technion's faculty of mathematics. His research focuses on development and theoretical analysis of algorithms, using mathematical tools from fields such as approximation theory, optimization, or invariant theory. Nadav has a Bsc and Msc in mathematics from the Hebrew University, and a PhD in applied mathematics and computer science from the Wiezmann institute. He was a postdoc in the Duke University math department in the years 2018-2021.