Farid Arthaud

Présentation

Photo of Farid

I'm currently a third-year PhD student at MIT in the EECS department, where I am advised by Martin Rinard. Previously, I was a student at École Normale Supérieure in Paris.

I am broadly interested in the theoretical analysis of multi-agent problems through the lens of game theory and learning theory. These topics lay at the intersection of several domains — theoretical computer science, probabilities, machine learning, statistics, combinatorics.

Non-academically, I'm passionate about trains, I like hiking in the Alps, visiting the most rural parts of France a train will take you. I'm interested in French geography and the science of climate change.

Contact: faais92jdrto arobase csail.mit.asdlkj.edu

Links: dblp | Scholar

Publications

Published

Preprints

Service

Reviewer: Journal of Global Optimization 2025, CogSci 2025.

Teaching: 6.1120 Dynamic computer language engineering (MIT Fall 2025).

Résumé

Éducation

2022- | MIT, Cambridge
PhD in Computer Science, advised by Martin Rinard.
2021-2022 | Université Paris-Saclay, Orsay
Research master in probabilities and statistics (Mathématiques de l'aléatoire), with a focus on statistics and machine learning.
2018-2022 | École Normale Supérieure (ENS), Paris
Bachelor degrees in computer science and mathematics, followed by the first year of masters in computer science.
Advised by Francis Bach and Pierre Gaillard.
2016-2018 | Lycée Champollion, Grenoble
Preparatory classes for grandes écoles.

Internships

2022 |

Approximation algorithms for correlation clustering using the Sherali-Adams hierarchy


Claire Mathieu and Vincent Cohen-Addad, IRIF, Université de Paris, Paris
Analysis of best-known upper bounds on approximation ratios for the correlation clustering problem.
2021 |

Theoretical guarantees for purely uniform random forests


Sylvain Arlot, Mathematics department, Université Paris-Saclay, Orsay
Analysis of point-wise, expected and integrated risk for purely uniform random forests.
2020 |

Delay predictions on a train network


SNCF Réseau, Paris, under the academic supervision of Vianney Perchet (ENSAE)
Proposal and implementation of a machine learning algorithm to predict trains' delays on a network, accounting for delay propagation along the graph.
2020 |

Few-short learning for machine translation in low-resource scenarios


Alexandra Birch and Rachel Bawden, School of Informatics, University of Edinburgh
Analysis of a novel technique to learn new vocabulary from very few ground-truth examples, using a measure of similarity in context between sentences.
2019 |

Theoretical guarantees for hierarchical clustering algorithms


Grigory Yaroslavtsev, School of Informatics, University of Indiana
Theoretical analysis of worst-case and average performance of hierarchical clustering algorithms for several different measures.

Projects

Fake French cities

Fascinated by the diversity of city names in France, I created an NLP model to create fake French city names using real French city names' phonemes. The website makes you discriminate between a fake and a real French city name, and it's harder than it seems.

Hierarchical clustering

Report (pdf)
My Bachelor's thesis presenting several important results for hierarchical clustering algorithms.

Work done with Claire Mathieu and Vincent Cohen-Addad.

Automata wallpapers

Gitlab
Tool written in OCaml to generate beautiful visualizations of cellular automata. The images on this page are generated using this tool.

If you also like these images, you can see the full-size images here:
1.png | 2.png | 3.png | 4.png | 5.png | 6.png | 7.png | 8.png | 9.png