Elisabetta Cornacchia

I am an Assistant Professor in the Department of Computing Sciences at Bocconi University (Milan, Italy). Previously I held postdoctoral positions at MIT, working with Prof. Elchanan Mossel, and at Inria and École Normale Supérieure (ENS) Paris, working with Prof. Laurent Massoulié. I received my PhD in Mathematics from EPFL (Lausanne, Switzerland), where I was advised by Prof. Emmanuel Abbé.

Research Interests: Theory of neural networks, high-dimensional statistics.

Here is a short CV.

Email: elisabetta [dot] cornacchia [at] unibocconi [dot] it

Youth in High Dimensions 2026:

Together with Jean Barbier (ICTP), Sebastian Goldt (SISSA) and Marco Mondelli (ISTA), we are organizing a seventh edition of Youth in High Dimensions, taking place on July 20th-23rd at ICTP in Trieste.

Our goal is to give emerging researchers the opportunity to gather and present their results, focusing on high-dimensional problems arising in deep learning, inference, statistical physics, and neuroscience.

Registration is free but mandatory, and can be completed on our website by 18 April 2026 for applicants requesting financial and/or visa support, and by 9 May 2026 for all other applicants.

We have a great line-up of speakers! All participants will have the opportunity to present a poster, and a limited number of slots will be available for contributed talks. See you in Trieste in July!

Publications:

M. Medvedev, I. Attias, E. Cornacchia, T. Misiakiewicz, G. Vardi, N. Srebro, Positive Distribution Shift as a Framework for Understanding Tractable Learning. Preprint, 2026. [arXiv]
E. Cornacchia, D. Mikulincer, E. Mossel, Low-Dimensional Functions are Efficiently Learnable under Randomly Biased Distributions. COLT 2025. [arXiv]
E. Abbe, E. Cornacchia, J. Hązła, D. Kougang-Yombi, Learning High-Degree Parities: The Crucial Role of the Initialization. ICLR 2025. [arXiv]
F. Bach, E. Cornacchia, L. Pesce, G. Piccioli, Theory and Applications of the Sum-Of-Squares Technique (Les Houches 2022 Lecture Notes). Journal of Statistical Mechanics: Theory and Experiments 2024. [arXiv]
E. Abbe, E. Cornacchia, A. Lotfi. Provable Avantage of Curriculum Learning on Parity Targets with Mixed Inputs. NeurIPS 2023. [arXiv ]
E. Cornacchia, E. Mossel. A Mathematical Model for Curriculum Learning for Parities. ICML 2023. [arXiv]
E. Abbe, S. Bengio, E. Cornacchia, J. Kleinberg, A. Lotfi, M. Raghu, C. Zhang. Learning to reason with neural networks: Generalization, unseen data and Boolean measures. NeurIPS 2022. [arXiv]
E. Abbe, E. Cornacchia, J. Hązła, C. Marquis. An initial alignment between neural network and target is needed for gradient descent to learn. ICML 2022. [arXiv]
E. Cornacchia*, F. Mignacco*, R. Veiga*, C. Gerbelot, B. Loureiro, L. Zdeborova. Learning curves for the multi-class teacher-student perceptron. Machine Learning: Science and Technology, 2022. [arXiv]
E. Abbe, E. Cornacchia, Y. Gu, Y. Polyanskiy. Stochastic block model entropy and broadcasting on trees with survey. COLT 2021 Best Student Paper Award. [arXiv]
E. Cornacchia, J. Hązła. Intransitive dice tournament is not quasirandom. Journal of Combinatorial Theory, 2020 [arXiv, Quanta article].
E. Cornacchia*, N. Singer, E. Abbe. Polarization in attraction-repulsion models. ISIT 2020. [arXiv]

*: denotes equally contributing first authors. In other papers, authors are listed in alphabetical order.

News article:

Quanta Magazine: Mathematicians Roll Dice and Get Rock-Paper-Scissors