معرفی
Omar Montasser is an Assistant Professor in the Department of Statistics and Data Science at Yale University. His research focuses on theoretical foundations of machine learning with emphasis on adversarial robustness and distribution shifts, addressing critical challenges in modern AI reliability.
His educational background includes:
- PhD in Computer Science from Toyota Technological Institute at Chicago (advised by Nathan Srebro)
- Combined BS/MS in Computer Science and Engineering from Penn State University (worked with Daniel Kifer and Sean Hallgren)
Montasser's research program centers on developing theoretical frameworks for robust machine learning. He investigates PAC learnability under adversarial perturbations, transformation invariances, and distribution shifts, bridging gaps between theoretical guarantees and practical robustness. His work provides foundational insights into why deep learning models fail under distributional changes and how to design provably robust systems.
Analysis of his publications from 2017-2025 reveals consistent focus on adversarial robustness theory. Key themes include minimax optimal learners for adversarial settings, OOD generalization guarantees, and computational aspects of robust learning. His most influential contributions characterize conditions for adversarially robust learnability and develop efficient algorithms for robust classification.
Scientific recognition includes:
- FODSI-Simons Postdoctoral Fellowship
- COLT 2019 Best Student Paper Award
- Multiple oral presentations at NeurIPS (2022)
- Spotlight presentation at NeurIPS 2020
He actively recruits PhD students for his research group, emphasizing theoretical machine learning. Teaching experience includes Information and Coding Theory (Fall 2022) and Learning Theory (TA, Fall 2020). While specific grants aren't detailed, his publication record in top venues (NeurIPS, COLT, ICML) suggests substantial research funding. He maintains active collaborations with leading researchers including Avrim Blum and Nathan Srebro.
Montasser leads a research group at Yale focused on theoretical machine learning, with current projects examining relaxed benchmarks in online classification and score matching in diffusion models. The group operates within Yale's Department of Statistics and Data Science, leveraging university resources for theoretical and computational research.
