Research

Here is a full list of my publications, thesis and project reports.
For published articles, visit my scholar page.

Publications

A. Azize and D. Basu, Some Targets Are Harder to Identify than Others: Quantifying the Target-dependent Membership Leakage. Primary version: TPDP2024 arxiv. In Proc. AISTATS2025 (Oral)
A. Azize and D. Basu, Open Problem: What is the Complexity of Joint Differential Privacy in Linear Contextual Bandits? In Proc. COLT2024
A. Azize, M. Jourdan, A. Al Marjani, and D. Basu. On the Complexity of Differentially Private Best-Arm Identification with Fixed Confidence. Primary version: EWRL2023. In Proc. Neurips2023
A. Azize and D. Basu, Concentrated Differential Privacy for Bandits. Primary version: In EWRL2023. In Proc. IEEE@SaTML2024
A. Azize and D. Basu, Rényi Differentially Private Bandits. In PPAI@AAAI, 2023 (PDF)
A. Azize and D. Basu, When Privacy Meets Partial Information: A Refined Analysis of Differentially Private Bandits. Primary version: EWRL2022. In Proc. Neurips2022
A. Azize and O. Gaizi, Conservative Optimistic Policy Optimization via Multiple Importance Sampling, arxiv

Invited Talks and Posters

2025:

• FairPlay Seminar, Criteo, Paris.
  

2024:

• SaTML Talk and Poster Session, University of Toronto, Canada.
  

• COLT, University of Alberta, Canada.
  

• Shimo Lab Seminar, University of Kyoto, Japan.
  

• IBIS Poster Session, Tokyo, Japan.
  

2023:

• Scool Seminar, Inria Lille, France.
  

• Comete Ethical AI Workshop, LIX, Ecole Polytechnique, France.
  

• Neurips Poster Session, New Orleans, US.
  

• EWRL Poster Session, Brussels, Belgium.
  

2022:

• Scool Seminar, Inria Lille, France.
  

• Neurips Poster Session, New Orleans, US.
  

• EWRL Poster Session, Politecnico di Milano, Italy.
  

• NeurIPS @ Paris Talk and Poster Session, Paris, France.
  

Reviewing

NeurIPS (2023-2024-2025), ICML (2022, 2023, 2024), COLT (2025), EWRL (2022, 2023), AAAI (2023), AISTATS (2025, Best Reviewer Award), TMLR

Open Problems

Better privacy guarantees for larger groups, (PDF)

Thesis

A. Azize, Privacy-Utility Trade-offs in Sequential Decision-Making under Uncertainty, PhD Thesis, manuscript, slides
A. Azize, Multi-Object Manipulation using Relational Reinforcement Learning, Msc Thesis, report

Project reports

A. Azize, Hard exploration for maze solving, github page, report
A. Azize and O. Gaizi, Efficiency of transmitted information in simple communicative agents, report