Research interests
- Recommender systems
- Multimodal and sequential learning
- Fair representation learning
- Optimisation and learning objectives
- Privacy-preserving machine learning
- Robust evaluation under distribution shift
About
Research profile
Bucher Sahyouni is a PhD candidate in Artificial Intelligence at the University of Surrey who passed his viva with minor corrections in September 2026. His research focuses on architecture and objective design for robust recommendation under sparse implicit feedback.
Current focus
His EPSRC-funded PhD examines how recommender architectures and objectives can extract stronger training signals from sparse, implicit interactions. The work spans multimodal and sequential recommendation, contrastive graph learning, competition-aware softmax objectives, and fairness-aware representation learning.
Research practice
His work covers the full research workflow: literature review, problem formulation, implementation, experimentation, ablation studies, analysis, visualisation, and paper writing. He has also reviewed for SIGIR, AAAI, ICML, ICLR, and RecSys.
Before the PhD, he completed a First-Class MEng in Electronic Engineering with Nanotechnology at the University of Surrey, building a foundation in algorithms, C++, control, communications, and electronic systems.