I am a fourth-year PhD student at ETH Zürich, under the supervision of Thomas Hofmann. My research focuses on understanding large-scale modern neural networks, from initialization to training dynamics. I have been recently awarded a Google Fellowship. Prior to my PhD, I earned a Master’s degree in Data Science from ETH Zürich and a Bachelor’s degree from Politecnico di Milano. I also interned at Amazon in Luxembourg and spent two years living in Shanghai.

News & Updates 📰

[December 2024] Visiting Princeton.

I will visit Boris Hanin at Princeton ORFE during the next semester. I am very grateful for this opportunity!

[September 2024] Google PhD Fellowship.

I feel honored to have been awarded a Google PhD Fellowship that will generously support me for the next year. Thank you Google!

[September 2024] Two papers accepted at Neurips 2024.

Co-authors and I had two papers accepted at Neurips. In Super Consistency of Neural Network Landscapes we find out that the curvature of the landscape is independent of scale (i.e. the landscape “looks the same”, no matter the scale). In Understanding and Minimising Outlier Features in Neural Network Training study the emergence of big outliers in LLM pre-training, and how modification of the architecture and optimizer can help!