Postdoctoral researcher @ Bocconi Institute for Data Science and Analytics. Interested in AI, physics, and a bunch of other things I need help keeping track of.
The London Geometry and Machine Learning summer school will take place in almost two months. If you wish to partner with us and sponsor the event, shoot me a DM or just complete the form in the post below. This is a fantastic opportunity!
LOGML will be back in July 2026 for its 5th edition! If you are working in Geometric Deep Learning, consider applying as a Mentor! It is an extremely interesting opportunity for young researchers to both lead a project and grow their network. Hoping to see many of you in London
π LOGML26β Mentor applications!
π Imperial College London | 13β17 July 2026
Mentor a project team. Attend talks, socials, networking events. πΈ Travel support available.
Deadline: 18 March 2026(AoE)
Apply: https://t.co/LnuzLuFBc2
#LOGML#MachineLearning#SummerSchool
Had a great time this afternoon at the Young Applied Mathematicians Conference, presenting my latest work. Fantastic audience and great talks throughout the day, especially regarding scientific ML.
About a year ago, I noticed that GNN-based SAT solvers hit a wall on tough problems, especially as the number of constraints increases. Today, I'm happy to share a possible theoretical explanation of this phenomenon, inspired by geometry and physics:
π https://t.co/1Q1fkWyUaz
Deadline approaching! Register by April 30 to join #RockinAI β the conference bridging statistical physics and machine learning, set in stunning Roccella, South Italy.
https://t.co/2EAvvNAEVW
We are pleased to announce the StatPhys29 Satellite Workshop "Learning and Optimization in High Dimensions," which will take place at Bocconi University, Milan, from July 9 to 11, 2025.
Further details can be found here https://t.co/hwR2a1kBV9
I had the pleasure of holding my first-ever guest lecture today, in a PhD course nonetheless. Huge thanks to Prof. @carlolucibello for inviting me. It was lovely to share this exciting view of deep learning with aspiring researchers. Funny how I was in their shoes not long ago...
Today, I had the pleasure to participate in a fantastic seminar by Prof. Yann Brenier, one of the pioneers of Optimal Transport. The topic was the relation between OT and Einstein's equations (in vacuum), giving an alternative view of gravitation (in terms of math frameworks)