Matteo Russo

Postdoctoral Researcher · Chair of Discrete Optimization & Theory Group, EPFL

prof_pic.jpg

EPFL

Lausanne, Switzerland

firstname dot lastname at epfl dot ch

about

I am a postdoctoral researcher in the Chair of Discrete Optimization and the Theory Group at EPFL, hosted by Friedrich Eisenbrand.

I completed my PhD in Data Science at Sapienza University of Rome in 2026, where I was fortunate to be advised by Stefano Leonardi.

Before that, I completed my Master’s degree in Computer Science at ETH Zurich in 2022 and my Bachelor’s degree in Computer Science at Princeton University in 2020.

research

My research in theoretical computer science lies at the intersection of Online Algorithms, Online Learning, and Optimization. I focus on sequential decision-making under uncertainty, and specifically on understanding which pieces of information are sufficient to overcome uncertainty, how their algorithmic value can be quantified, and when they are fundamentally insufficient. My work approaches this question through three main lenses:

  • Actively acquired information: how algorithms should choose what to observe when additional feedback, probes, comparisons, or queries are limited, costly, or noisy.
  • Weak stochastic information: how samples, moments, partial randomness, and other limited distributional information can improve sequential decisions while remaining robust to ambiguity and misspecification.
  • Geometric structure: how the geometry of the problem can constrain unseen inputs and act as substitute for information about the future.

I am also interested in applying these techniques to Algorithmic Game Theory, and related economic decision-making problems.

news

  • I am excited to have received the 2026 Best PhD Dissertation Award from the EATCS Italian Chapter!
  • Our paper Contextual Online Bilateral Trade has been accepted for publication at EC 2026!
  • Our paper Online Convex Optimization with Sublinear Noisy Probes has been accepted for publication at COLT 2026!

selected publications

  1. Nearly-Tight Bounds for Zonotope Containment and Beyond
    Friedrich Eisenbrand, Thomas Rothvoss, Matteo Russo, and Ruben Skorupinski
    Preprint
    arXiv
  2. Contextual Online Bilateral Trade
    Romain Cosson, Federico Fusco, Anupam Gupta, Stefano Leonardi, Renato Paes Leme, and Matteo Russo
    In ACM Conference on Economics and Computation, 2026
    arXiv
  3. Online Convex Optimization with Sublinear Noisy Probes
    Simone Di Gregorio, Anupam Gupta, Stefano Leonardi, and Matteo Russo
    In Conference on Learning Theory, 2026
    arXiv
  4. A Learning Perspective on Random-Order Covering Problems
    Anupam Gupta, Marco Molinaro, and Matteo Russo
    In SIAM Symposium on Simplicity in Algorithms, 2026
    arXiv
  5. Contract Design Beyond Hidden Actions
    Tomer Ezra, Stefano Leonardi, and Matteo Russo
    In ACM–SIAM Symposium on Discrete Algorithms, 2026
    arXiv
  6. Online Learning in the Random Order Model
    Martino Bernasconi, Andrea Celli, Riccardo Colini Baldeschi, Federico Fusco, Stefano Leonardi, and Matteo Russo
    In International Conference on Machine Learning, 2025
    arXiv
  7. A Tight VC-Dimension Analysis of Clustering Coresets with Applications
    Vincent Cohen-Addad, Andrew Draganov, Matteo Russo, David Saulpic, and Chris Schwiegelshohn
    In ACM–SIAM Symposium on Discrete Algorithms, 2025
    arXiv