Fabian Ruehle

About Me
Fabian Ruehle
email inspire google scholar github
Intro

I am a Theoretical Physicist and Computer Scientist interested in String Theory and its connections with Mathematics and AI. I have applied transformers and reinforcement learning to problems in knot theory and low-dimensional topology, Physics-informed Neural Networks (PINNs) as well as Kernel methods (including the Neural Tangent Kernel) to approximate Calabi-Yau metrics, Decision Trees to Sheaf Cohomology computations, as well as classical Neural Networks in supervised ML to learn topological data of Calabi-Yau manifolds. We also proposed Kolmogorov-Arnold Networks (KANs) as an alternative to NNs (see my Research for a summary).

I have also applied a variety of supervised and unsupervised ML tools (Regression, Decision Trees, Neural Networks, Support Vector Machines, Principal Component Analysis, Clustering Algorithms) to study a multi-omic blood atlas of Covid 19 patients.

Finally, I have written a Physics Report on Data science applications to string theory. It comes with a Github repo.

Seminars

I am involved in the organization of the following seminars (all times are Eastern time):

Tue11AM-12PMBi-weekly String Pheno seminar series (Zoom)
Tue12:30PM-1:30PMHigh Energy Physics Theory Seminar (Dana 114)
Fri12PM-1PM Math-Phys Lunch Seminar (Nightingale 544)
In the press

A collection of articles that have been written about my research: