Tennenbaum Early Career Professor and
Assistant Professor
Education
- Ph.D. Operations Research (2024), MIT
- M.Sc. Applied Mathematics (2020), École Polytechnique
- B.Sc. Applied Mathematics (2020), École Polytechnique
About
Moïse Blanchard is a Tennenbaum Early Career Professor and Assistant Professor at the H. Milton Stewart School of Industrial and Systems Engineering at Georgia Tech. Prior to Georgia Tech, he was a postdoctoral fellow at the Columbia University Data Science Institute (DSI) and received his Ph.D. degree at the Operations Research Center and the Laboratory for Information and Decision Systems at MIT, advised by Prof. Patrick Jaillet. Before his Ph.D., he graduated as the valedictorian of Ecole Polytechnique in 2019 with an M.S. degree in Applied Mathematics, and a B.S. degree in Mathematics, Computer Science and Physics.
Research
Moïse Blanchard's research interests broadly lie at the intersection of learnability in machine learning and statistics, online algorithms and optimization in decision-making. His research studies the fundamental limits of learning, investigating core tradeoffs between generalizability, data uncertainty, and computational resources. On the statistical side, he investigates how to designs algorithms that perform reliably under minimal data assumptions, quantifying how learning rates change as data becomes more challenging. On the computational side, he characterizes how constraints such as memory, communication, and feedback fundamentally limit algorithmic efficiency in classical optimization or statistical problems. This work provides principled foundations for designing and selecting learning algorithms that balance robustness, performance, and resource efficiency in real-world decision-making systems.
Representative Publications
- Moïse Blanchard. "Agnostic smoothed online learning", Symposium on Theory of Computing (STOC), 2025
- Yan Dai, Moïse Blanchard, Patrick Jaillet. "Non-Monetary Mechanism Design without Distributional Information: Using Scarce Audits Wisely". Conference on Learning Theory (COLT), 2025
- Moïse Blanchard. "Gradient Descent is Pareto-Optimal in the Oracle Complexity and Memory Tradeoff for Feasibility Problems", Symposium on Foundations of Computer Science (FOCS), 2024
- Moïse Blanchard, Patrick Jaillet. "Universal regression with adversarial responses". Annals of Statistics, 2023
- Moïse Blanchard, Junhui Zhang, Patrick Jaillet. "Quadratic memory is necessary for optimal query complexity in convex optimization: Center-of-mass is Pareto-optimal", Mathematics of Operations Research, 2023
- Moïse Blanchard, Adam Q. Jaffe. "Fréchet mean set estimation in the Hausdorff metric", via relaxation. Bernoulli, 2023
- Moïse Blanchard. "Universal online learning: An optimistically universal learning rule", Conference on Learning Theory (COLT), 2022
- Moïse Blanchard, Alexandre Jacquillat, Patrick Jaillet. "Probabilistic bounds on the k-Traveling Salesman Problem and the Traveling Repairman Problem". Mathematics of Operations Research, 2022
- Moïse Blanchard, Romain Cosson, Steve Hanneke. "Universal online learning with unbounded losses: Memory is all you need". International Conference on Algorithmic Learning Theory (ALT), 2022
- Moïse Blanchard, Jesús A. De Loera, Quentin Louveaux. "On the length of monotone paths in polyhedra". SIAM Journal on Discrete Mathematics, 2021