Salah-Eddine Rizki

Paris region, France

I am a PhD candidate in artificial intelligence at ETIS Laboratory, ENSEA, and CY Cergy Paris University, advised by Vassilis Christophides. My research focuses on graph representation learning and geometric deep learning, with current applications in molecular modeling and machine-learning interatomic potentials. I am also interested in generative modeling and explainable AI.

Education

PhD in Artificial Intelligence · CY Cergy Paris University, ETIS Laboratory
2025–2029 (expected) · Research on explainable graph diffusion models

Research Master’s in Data Science and Machine Learning · CY Cergy Paris University
2024–2025 · GPA: 3.35/4.00

Engineering Degree in Signal Processing and Artificial Intelligence · ENSEA
2022–2025 · GPA: 3.55/4.00

news

Jun 15, 2026	Presented a poster at the Learning on Graphs Paris Meetup 2026.
Mar 30, 2026	Gave the talk “Accelerating Science with AI” with Vassilis Christophides.

latest posts

Aug 23, 2026	Message-Passing Graph Neural Networks from Scratch: GCN and GAT in Pure PyTorch

selected publications

Counterfactual Explanations for Equivariant Graph Neural Networks: CE-EGNN for Molecular Property Regression

Salah-Eddine Rizki, Fatia Lekbour, and Guillaume Renton

In EDBT/ICDT 2026 Workshops, 2026

Bib PDF

@inproceedings{rizki2026ceegnn,
  title = {Counterfactual Explanations for Equivariant Graph Neural Networks: {CE-EGNN} for Molecular Property Regression},
  author = {Rizki, Salah-Eddine and Lekbour, Fatia and Renton, Guillaume},
  booktitle = {EDBT/ICDT 2026 Workshops},
  year = {2026},
}