Mame Diarra Toure
I am a PhD candidate at McGill University being advised by David Stephens. My research focuses on Bayesian deep learning and uncertainty quantification.
I develop scalable approaches that allow neural networks to model uncertainty effectively,
improving their robustness and interpretability. By exploring efficient variational methods and
principled ways of handling model complexity, I aim to make AI systems not only powerful but also
trustworthy. In 2024, I received the Women in AI Excellence Scholarship from
MILA.
Prior to joining McGill I worked as a quantitative analyst at Société Général. I did my undergraduate and part of my graduate training in France where I attended Lycée Sainte-Croix Sainte-Euverte and ENSIIE-France's Grandes Ecoles system. I hold an engineering degree in applied mathematics and computer science from ENSIIE and spent the third year of ENSIIE's curriculum at Paris Saclay University where I earned a Master in quantitative finance with first class honors. My outstanding academic performance was recognized with the award of the Sophie Germain Excellence Scholarship for my Master's in Quantitative Finance.
Email  / 
CV  / 
LinkedIn  / 
Twitter  / 
Github
|
|
Research
I’m interested in uncertainty quantification and
Bayesian deep learning, with a focus on scalable methods that
make AI systems more trustworthy.
|
Experience
Graduate Teaching Assistant, McGill University (Fall 2025)
Teaching assistant for MATH 139: Calculus with Precalculus in the Department of Mathematics and Statistics.
Responsibilities include leading tutorials, holding office hours, and supporting students on algebra, functions, limits,
and introductory calculus.
Worked on applied AI projects in natural language processing, including ingredient prediction for a food delivery company,
a matchmaking algorithm for startups and investors, and budget pacing for an ad company. Focused on topic modeling and
collaborative filtering methods.
Quantitative Analyst, Société Générale (2021–2022)
Contributed to the Haussmann Project, redesigning internal PD and LGD models. Proposed an approach to compute
default probabilities in low-default portfolios.
Quantitative Research Intern, BNP Paribas (2021)
Developed a mathematical model to quantify bank exposure to clearing houses. Strengthened problem formulation,
methodology development, and validation skills.
Research Intern, LaMME – Université d’Évry (Undergrad) (2019)
Investigated rough volatility models, with emphasis on kernel estimation for Volterra processes. Gained resilience and
adaptability through early research challenges.
Achievements
- Sophie Germain Excellence Scholarship — Awarded by FMJH for outstanding performance in mathematics.
- MILA Women in AI Excellence Scholarship (2024) — Recognized for academic excellence, leadership, and commitment to trustworthy AI.
Selected Projects
A few projects I completed during my undergraduate and graduate studies.
Mame Diarra Toure, Ghada Ben Said, Gabriel Moran, Ouassim Sebbar,
Houssem Fendi, Issame Sarroukhe
Using deep neural networks to approximate credit valuation adjustment (CVA).
Mame Diarra Toure, Imane Alla
Building ANN models to learn option price surfaces.
Mame Diarra Toure
Estimating the kernel in rough volatility models via Volterra process techniques.
Mame Diarra Toure
Study of the lifted Heston approach for capturing rough volatility behavior.
|