"With four parameters I can fit an elephant, and with five I can make him wiggle his trunk." - John von Neumann
- fedor.sergeev@ inf.ethz.ch
Department of Computer Science
Biomedical Informatics Group
- CAB F53
Informing machine learning models with human insights for better performance, reliability, and interpretability
I did my BSc in Applied Mathematics and Physics at MIPT. I developed computational methods for simulations in physics under the supervision of Igor Petrov and Nikolay Khoklov. In parallel, I worked on applications of deep learning in high-energy physics at GSI and LAMBDA.
In my MSc I studied Computational Sciences and Engineering EPFL, combining my interest in numerical and data-driven modeling. My thesis with Pascal Fua and Jonathan Donier was on physics-informed neural networks for modeling fluid flow. During my studies, I interned at startup companies Spiden and Neural Concept, working on synthetic data generation for medical spectroscopic data and 3D computer vision for advanced engineering, respectively.
I joined BMI lab in July 2023 to work on multimodal, representation and Bayesian deep learning on intensive care unit (ICU) data. I am also an ELLIS PhD student, co-supervised by Vincent Fortuin.