Bio

I obtained my Ph.D. in Mathematics and Computer Science from the Data Science and Mining Team at École Polytechnique, under the supervision of Prof. Michalis Vazirgiannis. In parallel, I work at BNP Paribas CIB. Based in Paris, I am part of the Data & AI Lab within the Quantitative Resarch team of Global Markets, where I conduct fundamental and applied research on graph machine learning for financial markets.

I also hold a Master’s degree in Applied Mathematics from École Polytechnique, a Master of Engineering in Image Processing and Data Science from Télécom Paris, and a Master of Engineering in Telecommunication and Computer Science from Lebanese University.

I am mainly interested in graph representation learning and signal processing. Currently, my work involves modelling the dynamic network of user–item interactions for various downstream tasks, such as market prediction and product recommendation.

---

News

• June 2025: I have successfully defended my PhD dissertation, entitled: Contributions to Graph Representation Learning and Applications in Financial Markets.
• June 2025: Our recent paper has been accepted in the Journal of Financial Data Science for the Summer Issue!
• September 2024: My second paper got accepted at ICAIF ‘24 and will be presented in New York!
• June 2023: My first Ph.D. paper got accepted and will be presented at RecSys ‘23 in Singapore.
• February 2023: I am teaching labs on “Machine Learning for Time Series” at École Polytechnique Executive Education.
• June 2022: I started my CIFRE Ph.D. at École Polytechnique & BNP Paribas CIB.
• October 2021: I defended my Master’s thesis on Generating Synthetic Financial Time Series with GANs.

---

Publications

2025

• Contributions to Graph Representation Learning and Applications in Financial Markets.
  Ashraf Ghiye
  [DOI]

• Graph-Based Factor Model for Interpretable Credit Spread Decomposition
  A. Ghiye, B. Barreau, L. Carlier, M. Vazirgiannis
  The Journal of Financial Data Science, Volume 7, Issue 3
  [DOI]

2024

• Rolling Forward: Enhancing LightGCN with Causal Graph Convolution for Credit Bond Recommendation
  A. Ghiye, B. Barreau, L. Carlier, M. Vazirgiannis
  5th ACM Conference on Artificial Intelligence for Finance (ICAIF 2024)
  [DOI] [arXiv]

2023

• Adaptive Collaborative Filtering with Personalized Time Decay Functions for Financial Product Recommendation
  A. Ghiye, B. Barreau, L. Carlier, M. Vazirgiannis
  17th ACM Conference on Recommender Systems (RecSys 2023)
  [DOI] [arXiv]