Pic at Dent de Jaman, 🇨🇭
Most of my time I spend as a PhD student of deep learning theory at the Physics of complex systems lab at EPFL.
I’m a machine learning experimentalist, my focus is in the rigorous study of existing deep neural networks to discover fundamental recurring laws that deepen our understanding of their unreasonable effectiveness. When deep nets get too complicated I resort to toy models 🐮 that reproduce the observed phenomena and allow for quantitative understanding.
A more serious summary of my research can be found under publications.
|Nov 23, 2022||How do CNNs aggregate spatial information? 🖼️ Check out our latest preprint here! 📜|
|Oct 23, 2022||After almost four months into running, I ran my first half-marathon in Lausanne! 🏃♂️🏅|
|Sep 15, 2022||Our latest paper on how feature learning can lead to overfitting in neural nets at NeurIPS 2022! ✨ See you in New Orleans!!|
|Feb 6, 2022||This summer you’ll find me at the Machine Learning Summer School in Krakow and at the first IAIFI Summer School in Boston! 👩🏫🤓|
|Sep 29, 2021||Our latest paper on diffeomorphisms stability in deep nets has been accepted at NeurIPS 2021!|