PhD Candidate — Technical University of Denmark
Stas
Syrota
Geometry & Bayesian probability in deep learning. Supervised by Søren Hauberg.
About
I am a PhD candidate at the Technical University of Denmark, supervised by Søren Hauberg. My work sits at the intersection of geometry, statistics, and deep learning — focused on the identifiability of latent variable models and Bayesian approaches to uncertainty in neural networks.
Before starting my PhD, I completed a Master’s in Mathematical Modeling at DTU and a Bachelor’s in Mathematics-Economics at the University of Copenhagen. In between, I worked as a Machine Learning Engineer building recommender systems and as a Data Scientist.
News
- Dec 2025 VIKING accepted at NeurIPS 2025
- Jul 2025 Paper on Identifying Metric Structures accepted at ICML 2025 (Vancouver)
- Jan 2025 Started PhD at the Technical University of Denmark
- Jul 2024 Decoder Ensembling presented at GRaM Workshop @ ICML 2024
Research
Writing
Photography
Coming soon
Coming soon
Coming soon
Coming soon