From Physics to Computer Science: Symmetry in Neural Networks with Prof. Tess Smidt
For the second episode we had the privilege of speaking with Professor Tess Smidt, Assistant Professor at MIT who specializes in the fields of Euclidean symmetry and neural networks. From her early days when she majored in physics and minored in architecture, to her current work in computer science, Tess has always sought to understand how different fields can inform one another.
A particularly fascinating aspect of our conversation was Tess's focus on symmetry. She argues that incorporating symmetry into neural networks can lead to more effective learning and better outcomes in scientific modeling.
Coming Up: Don't miss Prof. Smidt's presentation "Harnessing Symmetry and AI for Designing Physical Systems" on 14th May 2025, at Novotel in Kirchberg, Luxembourg.
Join us at Physics for Future on May 14th and 15th for two days of groundbreaking discussions and discoveries. Secure your spot now—registrations are open (until 30/04/2025) at the University of Luxembourg: https://www.uni.lu/fstm-en/conferences/physics-for-future/.
Sound: Chasing vampires by Victor_Natas -- https://freesound.org/s/694474/ -- License: Attribution 4.0
Physics for Future
Physics is our language to understand nature. It goes beyond formulas and allows us to shape the world. Hosted by Hanna Siemaszko, this podcast—brought to you by the Department of Physics and Materials Science (DPhyMS) at the University of Luxembourg—features conversations with some of the brightest minds in physics. You'll hear clear explanations about quantum mechanics, atomic clocks, laser cooling, AI, and data science from experts who know how to make these topics understandable. Originally launched as a lead-up to the Physics for Future conference, this three-episode miniseries makes complex ideas accessible for students and the curious public.
Where can you listen?
