About

Recently I finished my Master’s in Artificial Intelligence at Vrije Universiteit Amsterdam. My thesis with Shujian Yu about multimodal learning for MRI data can be found Here. The project was sponsored by NIAA research institue and you can find more about it [Here](https://networkinstitute.org/projects/depression-detection-from-multimodal-neuroimaging-data-towards-explainability-and-generalization/. Currently I work with Leiden University Medical Center on a semi-supervised segmentation with foundational model embeddings (like UNI2).

Previously I obtained my Bachelor at Warsaw University of Technology and then worked at Huawei for around 2 years on deep learning models for Automatic Speech Recognition. If you’re interested in working together feel free to drop me an email at fjakub.off@gmail.com!

I’m interested in exploring efficient foundational models, particularly with application in medical field. I’m also interested in exploring different strategies for multimodal fusion, pretraining of natively multimodal models and information theoretical approaches to training DNNs.

You can find examples of my academic writing in the Projects section.