Publications

2026

  1. AISTATS
    On the Hardness of Auditing Model Properties Under Updates: Complexity of Property-Preserving Updates
    Ayoub Ajarra, and Debabrota Basu
    In The 29th International Conference on Artificial Intelligence and Statistics, 2026
    PDF

2025

  1. AAAI
    Active Fourier auditor for estimating distributional properties of ML models
    Ayoub Ajarra, Bishwamittra Ghosh, and Debabrota Basu
    In Proceedings of the AAAI Conference on Artificial Intelligence, 2025
    PDF

Preprints

2026

  1. Efficient and Robust Active Multi-Group Fairness Auditing with Bias Canvases.
    Ajarra Ayoub, and Basu Debabrota
    2026
  2. On Gaming Fairness: Lifelong Online Auditing under Adversarial Model Owners.
    Ajarra Ayoub, and Basu Debabrota
    2026

Invited Talks and Posters

2025
Workshop on Algorithmic Fairness, EWAF 2025, Eindhoven, The Netherlands.

2024
Workshop on Regulatable ML , NeurIPS2024, Vancouver, Canada.
Scool Seminar, Inria Lille, France.

2023
Scool Seminar, Inria Lille, France.
Comete Ethical AI Workshop, LIX, Ecole Polytechnique, France.
Conférence sur l’Apprentissage Automatique (CaP), Université de Lille, France.

Reviewing Service

  • ICML (2023, 2024, 2025)
  • NeurIPS (2025)