Abdel Ghani Labassi

Bio

I am currently pursuing a PhD in the Applied Mathematics and Statistics Department at Johns Hopkins University, Baltimore, Maryland, USA, advised by Mateo Díaz.

I obtained a MSc in Computer Science in the Department of Computer Science and Operations Research at Université de Montréal, Quebec, Canada, advised by Didier Chételat and Andrea Lodi , affiliated to the Canada Excellence Research Chair in Data-Science for Decision-Making. In addition, I obtained a double major BSc in Computer Science and Mathematics from the same department.

Research Interests

My research interests center on understanding how algebraic and geometric structure in computationally challenging optimization problems arising in data science can be leveraged to design robust, efficient algorithms with theoretical guarantees.

Publications/In the pipeline

• Preconditioned subgradient method for composite optimization: overparameterization and fast convergence
  with Mateo Díaz and Liwei Jiang (*), 2025. PDF
• Learning to Compare Nodes in Branch and Bound with Graph Neural Networks
  with Didier Chételat and Andrea Lodi, NeurIPS, 2022. PDF

Teaching

I have held teaching assistant (TA) and course developer (CD) positions for, inter alia, the courses listed below.

Johns Hopkins University

• AMS 661 Optimization in Finance (Fall 2022, Fall 2023, Fall 2024)
• AMS 663 Network Models (Spring 2023, Spring 2025, Spring 2026)
• AMS 662 Optimization for Data Science (Summer 2025; Course Developer)
• AMS 653 Mathematical Game Theory (Spring 2024)

Université de Montréal

• IFT2125 Introduction to Algorithms (Winter 2021, Winter 2022)
• IFT2015 Data Structures (Winter 2021)
• IFT2105 Introduction to Theoretical Computer Science (Fall 2019)
• IFT1065 Discrete Mathematics (Winter 2022)

Technical Reports

Coming soon.

Random

Check out my Random Photograph Generator. This simple web app displays beautiful photography from Picsum. The page automatically refreshes every 30 seconds to show a new random photograph each time!

Contact

If you would like to get in touch, feel free to reach out via email at [email protected]. Short emails preferred.