Skip to CV
Diaa

Curriculum vitae

Abdulrahman Diaa

PhD Candidate · University of Waterloo · Ontario, Canada

a2diaa@uwaterloo.ca abdulrahmandiaa.ca Google Scholar LinkedIn

Education

The American University in Cairo

Sep 2016 - Jun 2021 Cairo, Egypt

BSc in Mathematics

Sep 2017 - Jun 2021 · GPA: 4.0/4.0
Thesis
Lattice-based Cryptography and Fully Homomorphic Encryption: A Survey
Supervisor
Ahmed El-Guindy

BSc in Computer Engineering

Sep 2016 - Jun 2021 · GPA: 4.0/4.0
Thesis
Mixture of Experts for Human Activity Classification from Images
Supervisors
Ahmed El-Guindy, Cherif Salama, and Mohammed Moustafa

Awards

Interop Labs PhD Fellowship University of Waterloo
CA$110K
President's Graduate Scholarship University of Waterloo
CA$10K
Best Poster Award Cybersecurity and Privacy Institute Conference
CA$1K
David R. Cheriton Graduate Scholarship University of Waterloo
CA$40K
International Masters Award of Excellence University of Waterloo
CA$12.5K
Johnston International Entrance Scholarship University of Waterloo
CA$5K
Tarek Nour AUC Scholarship The American University in Cairo
US$100K

Research Experience

Student Researcher

Google

Apr 2026 - Dec 2026 Ontario, Canada
Topics
Private RAG and nearest-neighbor search
Reference
Sarvar Patel

Undergraduate Data Science Research Intern

Dell Technologies - Data Office Team

Feb 2020 - Jun 2021 Cairo, Egypt
Topics
Technology forecasting and distributed neural networks
Reference
Steve Todd

Undergraduate Researcher

The American University in Cairo

Sep 2019 - Jan 2020 Cairo, Egypt
Topic
Deep reinforcement learning for traffic control
Reference
Cherif Salama

Publications

ZipPIR: High-throughput Single-server PIR without Client-side Storage

35th USENIX Security Symposium · 2026

Rasoul Akhavan Mahdavi, Abdulrahman Diaa, Florian Kerschbaum

  • Earlier version presented at FHE.org Conference 2023

FastLloyd: Federated, Accurate, Secure, and Tunable k-Means Clustering with Differential Privacy

34th USENIX Security Symposium · 2025

Abdulrahman Diaa, Thomas Humphries, Florian Kerschbaum

Optimizing Adaptive Attacks against Watermarks for Language Models

42nd International Conference on Machine Learning (ICML) · 2025

Abdulrahman Diaa, Toluwani Aremu, Nils Lukas

  • Best Poster Award, CPI Annual Conference 2024
  • Oral Presentation (3.9%), WMARK@ICLR 2025
  • Spotlight Poster (2.6%), ICML 2025

Leveraging Optimization for Adaptive Attacks on Image Watermarks

12th International Conference on Learning Representations (ICLR) · 2024

Nils Lukas, Abdulrahman Diaa, Lucas Fenaux, Florian Kerschbaum

Fast and Private Inference of Deep Neural Networks by Co-designing Activation Functions

33rd USENIX Security Symposium · 2024

Abdulrahman Diaa, Lucas Fenaux, Thomas Humphries, Marian Dietz, Faezeh Ebrahimianghazani, Bailey Kacsmar, Xinda Li, Nils Lukas, Rasoul Akhavan Mahdavi, Simon Oya, Ehsan Amjadian, Florian Kerschbaum

Privacy-Preserving Federated Recurrent Neural Networks

Proceedings on Privacy Enhancing Technologies (PoPETs) 2023 · 2023

Sinem Sav, Abdulrahman Diaa, Apostolos Pyrgelis, Jean-Philippe Bossuat, Jean-Pierre Hubaux

Patents

Splitting Neural Networks on Multiple Edge Devices to Train on Vertically Distributed Data

Dell Technologies

Mar 2024
Patent
US 11934487B2 (Granted)

Forecasting Technology Phases Using Unsupervised Clustering with Wardley Maps

Dell Technologies

Jan 2021
Application
US 17/160782 (Pending)

Academic Service

Reviewer 40th Annual Conference on Neural Information Processing Systems (NeurIPS)
Workshop Reviewer 1st Workshop on GenAI Watermarking (WMark@ICLR)
Artifact Reviewer (Ninja Award) 34th USENIX Security Symposium
Journal Reviewer Transactions on Knowledge and Data Engineering (TKDE)

Teaching Experience