Andrea Paudice

alt text 

Assistant Professor (Tenure Track)
Department of Computer Science, Aarhus University

Contact
Email: and [dot] paudice [at] gmail [dot] com
Others: Google Scholar, GitHub, LinkedIn

About me

I am assistant professor (tenure track) at the Department of Computer science of Aarhus University, where I'm part of the Algorithms, Data Structure, and Foundations of Machine Learning group.

Prior to this, I was a postdoctoral researcher at the University of Milan, Italy. I worked among the LAILA team led by professor Nicolò Cesa-Bianchi.

I obtained my PhD from the University of Milan, under the supervision of Nicolò Cesa-Bianchi and Massimiliano Pontil.

You can find more about my background on my LinkedIn.

Research Interests

My recent research interests revolve around the theoretical foundations of machine learning with emphasis on the impact of heavy-tailed noise.

More precisely my interests spans the following topics:

  • Stochastic optimization

  • Statistical learning theory and concentration inequalities

  • Applications to adversarial learning and computer security

My list of publications is available on my Google Scholar.

If you are interested in working with me feel free to drop me an email!

Openings

I'm currently looking for a PhD student with a solid mathematical background to work on topics in stochastic optimization and generalization bounds. If you are interested, drop me an email with a short description of your interests and a copy of your resume.

News

  • Sep. 2025: our work "Revisiting Agnostic Boosting" on statistical near-optimal agnostic boosting got accepted at NeurIPS (joint work with A. da Cunha, M.M. Høgsgaard, and Y. Sun)!

  • May 2025: our work "Uniform Mean Estimation for Heavy-Tailed Distributions via Median-of-Means" on uniform convergence under for heavy-tailed data got accepted at ICML (joint work with M.M. Høgsgaard)!

  • Nov. 2024: our work on fat-shattering dimension bounds got accepted at IPL (joint work with E. Esposito and R. Colomboni)!

  • Oct. 2024: our work on clipped subgradient methods got accepted at SIAM Journal on Mathematics of Data Science (joint work with D.A. Parletta, M. Pontil, and S. Salzo)!