Caterina Urban

Research Scientist (Chargé de Recherche)

I am a member of the INRIA research team ANTIQUE (ANalise StaTIQUE). Previously, I was a postdoctoral researcher at the Chair of Programming Methodology, led by Peter Müller at ETH Zurich. I completed my Ph.D. in Computer Science, working under the joint supervision of Radhia Cousot and Antoine Miné at École Normale Supérieure. I hold a Bachelor’s and a Master’s degree in Computer Science, both received with full marks and honors (summa cum laude) from the Università degli Studi di Udine.

My research interests span the whole spectrum of formal methods and aim at developing methods and tools to enhance the reliability of computer software and understanding complex software systems. My main area of expertise is static analysis based on abstract interpretation, which provides rigorous mathematical guarantees of the behavior of computer programs. I am currently engaged in a long-term research effort to enhance the understanding and reliability of data science software, which nowadays plays an increasingly important role in critical decision making in fields such as finance and medicine.

Schedule a meeting with me using Calendly

News

📑 New paper! Reduced Products of Abstract Domains for Fairness Certification of Neural Networks (co-authored with Denis Mazzucato) has been accepted to SAS 2021!

👥 I am co-chairing with Sandra Kiefer and Marijana Lazić the FLOC 2022 Mentoring Workshop!

👩🏼‍💻 I have been developing a Python interface for the APRON numerical abstract domain library. Contributions are welcome!

Interests

Static Program Analysis
Abstract Interpretation
Programming Languages
Data Science
Machine Learning

Education

PhD in Computer Science, 2015

École Normale Supérieure, Paris, France
MSc in Computer Science, 2011

Università degli Studi di Udine, Udine, Italy
BSc in Computer Science, 2009

Università degli Studi di Udine, Udine, Italy

Research Projects

FuncTion

The goal of the FuncTion project was the development of a static analyzer which automatically infers ranking functions and sufficient precondition for program termination (and other liveness properties) by means of abstract interpretation.

Libra

The goal of the Libra project is to develop new analyses and tools to reason about and certify fairness of decision-making software.

Lyra

The Lyra research project is a long-term research effort to enhance the understanding and reliabilty of data science software. It aims ad developing new practical and accessible analyses and tools to reason about and provide rigorous guarantees of the behavior of data analytics, big data, machine learning, and deep learning applications.

Recent Publications

More Publications

Denis Mazzucato, Caterina Urban. Reduced Products of Abstract Domains for Fairness Certification of Neural Networks. In SAS, 2021.

Code Project Artifact

Caterina Urban, Antoine Miné. A Review of Formal Methods applied to Machine Learning. CoRR abs/2104.02466, 2021.

PDF Project arXiv HAL

Francesco Ranzato, Caterina Urban, Marco Zanella. Fair Training of Decision Tree Classifiers. CoRR abs/2101.00909, 2021.

PDF Project arXiv

Recent Talks

More Talks

An Abstract Interpretation Recipe for Machine Learning Fairness

Sunday, July 18, 2021 9:00 AM

4th Workshop on Formal Methods for ML-Enabled Autonomous Systems (FoMLAS 2021), 🇺🇸 Los Angeles, USA (remote)

Code Project

Perfectly Parallel Fairness Certification of Neural Networks

Wednesday, May 5, 2021 5:00 PM

Séminaire du DI, 🇫🇷 École Normale Supérieure, France (remote)

PDF Code Project

Formal Methods for Robust Artificial Intelligence: State of the Art

Friday, February 5, 2021 2:00 PM

🇫🇷 Airbus, France (remote)

Slides Project

Students

Current Students

Denis Mazzucato (PhD, École Normale Supérieure: Oct 2020 -)

Former Students

Serge Durand (M1 Internship, École Normale Supérieure: Jun - Aug 2020)
ENS Paris-Saclay, France
Marco Zanella (PhD Internship, École Normale Supérieure: May - Aug 2020)
Università degli Studi di Padova, Italy
Radwa Sherif Abdelbar (Bachelor’s Thesis, ETH Zurich: Mar - Aug 2018)
German University in Cairo, Egypt
Lowis Engel (Master’s Thesis, ETH Zurich: Feb - Aug 2018)
ETH Zurich, Switzerland
Madelin Schumacher (Master’s Thesis, ETH Zurich: Sep 2017 - Mar 2018)
ETH Zurich, Switzerland
Samuel Ueltschi (Master’s Thesis, ETH Zurich: Mar - Sep 2017)
ETH Zurich, Switzerland
Mostafa Hassan (Bachelor’s Thesis, ETH Zurich: Mar - Aug 2017)
German University in Cairo, Egypt
Simon Wehrli (Master’s Thesis, ETH Zurich: Feb - Aug 2017)
ETH Zurich, Switzerland
Flurin Rindisbacher (Master’s Thesis, ETH Zurich: Mar - Aug 2017)
ETH Zurich, Switzerland
Severin Münger (Master’s Thesis, ETH Zurich: Sep 2016 - Mar 2017)
ETH Zurich, Switzerland
Nathanaël Courant (L3 Internship, ETH Zurich: Jun - Jul 2016)
École Normale Supérieure, France
Lukas Neukom (Master’s Thesis, ETH Zurich: Mar - Sep 2016)
ETH Zurich, Switzerland
Seraiah Walter (Master’s Thesis, ETH Zurich: Feb - Aug 2016)
ETH Zurich, Switzerland

Teaching

Year 2021-2022

Abstract Interpretation: Application to Verification and Static Analysis
M2, Master Parisien de Recherche en Informatique (MPRI), Université de Paris, France

Year 2020-2021

Abstract Interpretation and Applications Beyond the Beaten Track
PhD, Doctoral Program in Computer Science, Gran Sasso Science Institute (GSSI), Italy
Abstract Interpretation: Application to Verification and Static Analysis
M2, Master Parisien de Recherche en Informatique (MPRI), Université de Paris, France

Software

ApronPy

Python Interface for the APRON Numerical Abstract Domain Library

FuncTion

Abstract Interpretation-based Static Analysis for (Conditional) Termination (and Other CTL Properties)

Libra

Perfectly Parallel Abstract Interpretation-based Fairness Certification for Neural Networks

Lyra

Abstract Interpretation-based Static Analysis for Data Science Applications

Typpete

SMT-based Static Type Inference for Python 3.x

Contact

caterina.urban@inria.fr
Schedule a meeting with me using Calendly
ANTIQUE Research Team
École Normale Supérieure
45, rue d'Ulm
75005 Paris, France