Rafael Pinot
Research Teaching Outreach
Rafael Pinot

Research Interests


My main line of research is in statistical machine learning with a focus on the privacy and robustness of machine learning algorithms. Most of my work has a theoretical flavor, but I also like to work on more applied projects to get deeper understanding of state-of-the-art methods, or simply to better grasp the gap that can exist between the theoretical analysis and the empirical performance of some machine learning algorithms. Below you will find a list of my published work in books, journals and conferences. For a full (up to date) list of my work, please visit my google scholar.

Phd and Master Thesis

  • On the impact of randomization on robustness in machine learning
    PhD Thesis at PSL University in 2020
    Jury: Prof Jamal Atif, Prof. Francis Bach, Dr. Sébastien Bubeck, Prof. Stéphane Canu, Dr Cedric Gouy-Pailler, Dr. Panayotis Mertikopoulos, Prof. Cordelia Schmid, and Prof. Michèle Sebag
  • Minimum spanning tree release under differential privacy constraints
    Master Thesis at 2017 Sorbonne University
    Jury: Prof. Jamal Atif, Dr. Cédric Gouy-Pailler, Dr. Maxime Sagnier, and Dr. Florian Yger

Books, Book Chapters

  • Robust Machine Learning: Distributed Methods for Safe AI
    Book published by Springer Edition in 2024
    Authors: R Guerraoui, N Gupta, R Pinot
  • Large Language Models in Cybersecurity: Threats, Exposure and Mitigation
    Collective Book published by Springer Edition in 2024
    Editors: A Kucharavy, O Plancherel, V Mulder, A Mermoud, V Lenders

Journal Papers

  • Byzantine Machine Learning: A Primer
    ACM Computing Survey 2023
    Authors: R Guerraoui, N Gupta, R Pinot
  • On the robustness of randomized classifiers to adversarial examples
    Machine Learniong Journal 2022.
    Authors: R Pinot, L Meunier, F Yger, C Gouy-Pailler, Y Chevaleyre, J Atif
  • SPEED: Secure, PrivatE, and Efficient Deep learning
    Machine Learning Journal 2021
    Authors: A Grivet Sébert, R Pinot, M Zuber, C Gouy-Pailler, R Sirdey

Conference Papers

  • Towards Practical Homomorphic Aggregation in Byzantine-Resilient Distributed Learning
    International Middleware Conference (Middleware) 2024
    Authors: A Choffrut, R Guerraoui, R Pinot, R Sirdey, J Stephan, M Zuber
  • Fine-Tuning Personalization in Federated Learning to Mitigate Adversarial Clients
    Conference on Neural Information Processing Systems (NeurIPS) 2024
    Authors: Youssef Allouah, Abdellah El Mrini, Rachid Guerraoui, Nirupam Gupta, Rafael Pinot
  • Revisiting Ensembling in One-Shot Federated Learning
    Conference on Neural Information Processing Systems (NeurIPS) 2024
    Authors: Youssef Allouah, Akash Dhasade, Rachid Guerraoui, Nirupam Gupta, Anne-Marie Kermarrec, Rafael Pinot, Rafael Pires, Rishi Sharma
  • Tackling Byzantine Clients in Federated Learning
    International Conference on Machine Learning (ICML) 2024
    Authors: Y Allouah, S Farhadkhani, R Guerraoui, N Gupta, R Pinot, G Rizk, S Voitovych
  • Robust Distributed Learning: Tight Error Bounds and Breakdown Point
    Conference on Neural Information Processing Systems (NeurIPS) 2023
    Authors: Y Allouah, R Guerraoui, N Gupta, R Pinot, G Rizk
  • On the Inherent Anonymity of Gossiping
    International Symposium on Distributed Computing (DISC) 2023
    Authors: R Guerraoui, AM Kermarrec, A Kucherenko, R Pinot, S Voitovych
  • On the Privacy-Robustness-Utility Trilemma in Distributed Learning
    International Conference on Machine Learning (ICML) 2023
    Authors: Y Allouah, R Guerraoui, N Gupta, R Pinot, J Stephan
  • Robust Collaborative Learning with Linear Gradient Overhead
    International Conference on Machine Learning (ICML) 2023
    Authors: S Farhadkhani, R Guerraoui, N Gupta, L. Hoang, R Pinot, J Stephan
  • Fixing by Mixing: A Recipe for Optimal Byzantine ML under Heterogeneity
    Conference on Artificial Intelligence and Statistics (AISTATS) 2023
    Authors: Y Allouah, S Farhadkhani, R Guerraoui, N Gupta, R Pinot, J Stephan
  • Towards Consistency in Adversarial Classification
    Conference on Neural Information Processing Systems (NeurIPS) 2022
    Authors: L Meunier, R Ettedgui, R Pinot, Y Chevaleyre, J Atif
  • Byzantine Machine Learning Made Easy by Resilient Averaging of Momentums
    International Conference on Machine Learning (ICML) 2022
    Authors: S Farhadkhani, R Guerraoui, N Gupta, R Pinot, J Stephan
  • The Universal Gossip Fighter
    IEEE International Parallel & Distributed Processing Symposium (IPDPS) 2022
    Authors: A Gorbunova, R Guerraoui, AM Kermarrec, A Kucherenko, R Pinot
  • Mixed Nash Equilibria in the Adversarial Examples Game
    International Conference on Machine Learning (ICML) 2021
    Authors: L Meunier, M Scetbon, R Pinot, J Atif, Y Chevaleyre
  • Differential Privacy and Byzantine Resilience in SGD: Do They Add Up?
    ACM Symposium on Principles of Distributed Computing (PODC) 2021
    Authors: R Guerraoui, N Gupta, R Pinot, S Rouault, J Stephan
  • Randomization Matters. How to Defend against Strong Adversarial Examples
    International Conference on Machine Learning (ICML) 2020
    Authors: R Pinot, R Ettedgui, G Rizk, Y Chevaleyre, J Atif
  • Theoretical Evidence for Adversarial Robustness through Randomization
    Conference on Neural Information Processing Systems (NeurIPS) 2019
    Authors: R Pinot, L Meunier, A Araujo, H Kashima, F Yger, C Gouy-Pailler, J Atif
  • Graph-based Clustering under Differential Privacy
    Uncertainty in Artificial Intelligence (UAI) 2018
    Authors: R Pinot, A Morvan, F Yger, C Gouy-Pailler, J Atif