Angelo Gnazzo

Angelo Gnazzo

PhD @ TU Wien - AITHYRA

Angelo Gnazzo

TU Wien & AITHYRA PhD student.

I am a PhD student in Machine Learning at TU Wien and AITHYRA, advised by İsmail İlkan Ceylan and Michael Bronstein.

My research focuses on the development of fundamental methodology on generative models, with a particular interest in discrete generative models (language, graphs and other discrete structures) and also the understaing of their statistical guarantees and of their capacity for memorization, generalization, and reasoning. These questions are tightly coupled with fundamental challenges in AI4Science, where the ability to generalize beyond the training set and perform structured exploration is critical for large-scale scientific discovery.

Prior to my PhD, I spent some time in industry, working as a Quant Researcher for LMR Partners within the Inflation and Systematic Fixed Income Team, where I worked on predictive models for inflation and systematic strategies around auction cycles. I later joined MapleCap Partners as a founding team member following its spinoff from LMR.

I completed my MSc in Mathematics at ETH Zürich, where I was fortunate to be supervised by March Boedihardjo and Afonso Bandeira for my master thesis, titled “Covariance Estimation for High-Dimensional Distributions”.

My academic background is in high-dimensional probability and random matrix theory, and I remain deeply fascinated by what could be called “mathematical theory of big data and machine learning”.

Feel free to reach out if you wish to collaborate, exchange ideas, or seek thesis supervision.

news

  • March, 2026: We are teaching Graph Representation Learning at TU Wien this semester, where I am assisting İsmail as a Teaching Assistant.
  • February, 2026: I started my PhD at TU Wien and AITHYRA, under the supervison by İsmail İlkan Ceylan and Michael Bronstein.
  • September, 2025: I have attented the “Random Matrix Theory for Learning and Statistical Physics” Workshop.
  • April, 2024: MapleCap Partners, spinoff from LMR was officially launched and I am proud to be part of it.
  • September, 2023: I have joined LMR Partners in the Zürich offices to work as a Quant within the Inflation Team.
  • August, 2023: I have attented the Mediterranean Machine Learning Summer School 2023 edition, which took place in Thessaloniki.
  • July, 2023: I have defended my Master Thesis, supervised by March Boedihardjo and Afonso Bandeira and titled “Covariance Estimation for High-Dimensional Distributions”. We studied the problem of estimating covariance matrices. In particular, the main question we posed ourselves is: “Which is the minimal sample size n to guarantee that the sample covariance matrix approximates with epsilon accuracy the true covariance matrix?” Take a look at it here.
  • June, 2022: I started working at ETH AI Center as a research intern, advised by Amartya Sanyal. I am going to be focusing on problems at the intersection of Differential Privacy and Online Learning.
  • June, 2022: I finished my semester research project at ETH. I worked under the supervision of Nikita Zhivotovskiy and Afonso Bandeira on a survey paper, titled “Generalization Error Bound via Information-Theoretic Methods and Convex Analysis”. Take a look at it here.
  • March, 2022: I published my first blog post on Medium about the “Method of Interlacing Polynomials”, introduced by Adam Marcus, Daniel Spielman, Nikhil Srivastava. Take a look at it here. Hopefully, I will be posting more blog posts in the next few months.
  • September, 2021: I started my MSc in Mathematics at ETH Zurich.
  • July, 2021: I graduated from Politecnico of Milano, obtaining my BSc degree in Mathematical Engineering with the grade of 110/110 cum laude.