Chase van de Geijn

ML Physicist

About Me
Interests
Blogs
Research
Education
Teaching
Supervision

About Me

Profile Image

Welcome to my Theory of Everything!

Howdy! I'm Chase van de Geijn, a eccentric AI Researcher and Orange Enthusiast with an interest in understanding AI at a deep mathematical level. I am doing my PhD at the University of Göttingen mainly focusing on Foundation models for Neuroscience. In the past, I had started a PhD in Edinburgh in Applied Math target towards Geometric DL for Fluid Dynamics. I am also a co-organizer of the NeurReps Workshop in 2024 and 2025, but have stepped away to focus on teaching a GDL course this year.

Research Interests

My main specialty is in Geometric Deep Learning particularly Clifford Algebra. However, I also have done work in Bayesian Neural Networks, Wavelet Theory, AI4Histopathology and have recently gotten into Computational Neuroscience, particularly sparse coding, Vector Symbolic Architectures, and geometric perception/neurogeometry. Generally, I can be described as a highly opinionated, goofy guy, who loves to learn and teach.

Personal Interests

In my free time, I enjoy walking, baking, and board games. I am also an enthusiast of the color orange and Dungeons and Dragons. I am an avid fan of Dimension 20 and Dungeons and Dads.

Education

PhD
Master's Degree
Bachelor's Degree

Teaching

Courses
Guest Lectures

Supervision

Supervision

Interests

Clifford Algebra
Geometric Deep Learning
Fluid Dynamics

Blogs

RoPE
Long Context, Provably
RollPE

Research

Hierarchical Equivariance
Equivariant Neural Fields
Group Generalized POD
Cake Wavelets

Skills

Teaching

Teaching

I am passionate about teaching and take every opportunity to share my knowledge with others. I have experience as a teaching assistant in both Bachelor's and Master's level courses.

  • Autonomous Mobile Robots : UvA AI, Bachelors Level
  • Applied Machine Learning : UvA Datascience, Bachelors Level
  • Machine Learning 1 : UvA AI, Masters Level

I frequently give colloquium lectures about my research for various groups at the University of Edinburgh. I have given the following lectures:

  • Hierarchical Geometric Deep Learning : Pure Math for AI - Post Graduate Applied Math Colloquium , The University of Edinburgh, May 2024
  • Lifting to SE(2) should be a Piece of Cake - Machine Learning Reading Group, The University of Edinburgh, April 2024
  • Lifting to SE(2) should be a Piece of Cake - Redwood Center of Theoretical Neuroscience Berkeley, Dec 2023
  • Lifting to SE(2) should be a Piece of Cake - Machine Learning and Simulation Science Lab, University of Stuttgart Aug 2023
  • Learning the Schrodinger Equation with Uncertainty with Bayesian Neural Networks - AMLab, University of Amsterdam June 2019
  • Wavelet Theory for Signal Processing

Subtab Image