Nassim Arifette
PhD student (from October 2026) · École Polytechnique
I am starting a PhD at École Polytechnique in October 2026. My thesis studies RNA–RNA interactions, in particular in the influenza A virus, using machine-learning and computational methods.
Before that, I completed the MVA master’s programme at ENS Paris-Saclay and did research internships in 3D medical imaging (BioMaps, CEA / Université Paris-Saclay), neural-network verification (LIX, École Polytechnique), knowledge graphs (LISN), and computational phylogenetics (CIRB, Collège de France).
Projects
Hi! PARIS Hackathon 2025 – EduTrackWinner
2025Winner project on education built during the Hi! PARIS 2025 hackathon.
Versailles Concierge — Les 4 MousquetAIres4th place
2025A multilingual AI concierge that turns open questions into concrete day plans across the Versailles estate.
Unpaired 3D medical image translation with custom losses.
Comparative analysis of deep learning approaches for entity alignment in knowledge graphs (YAGO/Wikidata/DBpedia) at LISN, Univ. Paris-Saclay.
An exploratory study of abstraction-based ReLU verification using zones (DBMs), tropical geometry, and Julia prototypes.
Intégration d’une projection sur le cône métrique dans Phyloformer pour garantir la métricité des distances phylogénétiques.
MFCCs vs. Mel spectrograms for frame-level voiced/unvoiced detection with a small CNN—compact cepstra outperform higher-dimensional inputs on limited data.
Publications & reports
Unpaired 3D CT-to-MRI Synthesis with Histogram-Aware CycleGANs
N. Arifette · Research internship report, BioMaps (CEA / Université Paris-Saclay), 2025
Neural Network Verification via Set Analysis
N. Arifette · M1 research report, LIX, École Polytechnique, 2024
Projection convexe pour garantir l’inégalité triangulaire dans Phyloformer
N. Arifette · Bachelor’s research report, CIRB, Collège de France, 2023
Voiced/Unvoiced Classification in Speech with Small CNNs
N. Arifette · TER report, Université Paris-Saclay, 2023
Recent notes
An introductory overview of how category theory concepts can be applied to artificial intelligence.
A worked path through Seely's 2-categorical lambda calculus: from one term with two beta reductions to interchange, laxity, and the adjunction between currying and uncurrying.
A story-driven tour of new papers and benchmarks—plus what actually moved the needle in my pipelines, with links and practical takeaways.
How to wrap language models with tools, memory, and guardrails so they can pursue goals safely.
Contact
The best way to reach me is by email at nassim.ari@gmail.com.