University of Basel

Department of
Mathematics and Computer Science

Optimization of Machine Learning Systems Group

Search
Enea Monzio Compagnoni
Assistant / PhD candidate
Philosophisch-Naturwissenschaftliche Fakultät
Departement Mathematik & Informatik
FG Lucchi

Assistant / PhD candidate

Spiegelgasse 1
4051 Basel
Schweiz

enea.monziocompagnoni@unibas.ch

Bio

I am a second year PhD student at University of Basel working under the supervision of Aurelien Lucchi. Prior to that, I have worked three years at UBS as AI Quantitative Analyst. I obtained my first Master of Science in Mathematics from the University of Milan and my second Master of Science in Quantitative Finance from ETH Zürich.

I am interested in Stochastic Optimization for Deep Learning, Implicit Bias, and their impacts on real world applications.

Education

Industry Experience

  • Yahoo! Research | Scalable ML ‑ Intern | 07/23 – 10/23 | Munich, Germany
    • Supervisor: Dr. Strom Borman, Dr. Hans Kersting
    • Project 1: Theorized and Implemented a Risk-Aware Framework for Optimal Control of Advertisement Budgeting.
    • Project 2: Increased Performance of the Spending Control by 4% (A/B Testing Included).
  • UBS AG | AI Quantitative Analyst - Full Time | 07/19 – 10/22 | Zurich, Switzerland
    • Selected Project 1: Translated best practice of Credit Officers and Advisors into a concrete model for Lombard Lending to Ultra-High-Net-Worth Clients.
    • Selected Project 2: Increased accuracy of Liquidity forecasting by 3%: Coupled extensive model selection, statistical hypothesis testing, and business requirements.
    • Selected Project 3: Improved Anti-Money Laundering methodology: Developed classification and clustering models (e.g. AdaBoost, t-SNE). Found previously unknown patterns.

Publications

[4] Enea Monzio Compagnoni, Luca Biggio, Antonio Orvieto, Frank Norbert Proske, Hans Kersting, Aurelien Lucchi, An SDE for Modeling SAM: Theory and Insights, In Proc. of 40th International Conference on Machine Learning, 2023.

[3] Matteo Burzoni*, Alessandro Doldi*, Enea Monzio Compagnoni*, Risk Sharing with Deep Neural Networks, arXiv preprint arXiv: 2212.11752, 2023.

[2] Enea Monzio Compagnoni, Anna Scampicchio, Luca Biggio, Antonio Orvieto, Thomas Hofmann, Josef Teichmann, On the Effectiveness of Randomized Signatures as Reservoir for Learning Rough Dynamics, In Proc. of International Joint Conference on Neural Networks (IJCNN), 2023.

[1] Enea Monzio Compagnoni, Luca Biggio, Antonio Orvieto, Empirics on the Expressiveness of Randomized Signature, The Symbiosis of Deep Learning and Differential Equations: DLDE Workshop - NeurIPS 2021, 2021.

* denotes equal contributions.

To top