Ph.D. Student, Department of Computer Science, University of Toronto, mousavi [@] cs (dot) toronto (dot) edu [CV] [Google Scholar]

About me

I am a computer science Ph.D. student at the University of Toronto and Vector Institute, advised by Murat A. Erdogdu. Before that, I received my B.Sc. in computer engineering from Sharif University of Technology. During Fall 2023, I visited Lénaïc Chizat at the EPFL Institute of Mathematics. During Summer 2025, I am interning at Apple ML Research with Marco Cuturi.

Research Interests

I am broadly interested in statistical learning theory. At the moment, I am focusing on

• Deep Learning Theory: The dynamics of gradient-based training and representation learning

• Non-convex Sampling and Optimization: Obtaining complexity guarantees under functional inequalities

Publications

• When Do Transformers Outperform Feedforward and Recurrent Networks? A Statistical Perspective
  Alireza Mousavi-Hosseini, Clayton Sanford, Denny Wu, Murat A. Erdogdu.
  Preprint, 2025.

• Robust Feature Learning for Multi-Index Models in High Dimensions.
  Alireza Mousavi-Hosseini, Adel Javanmard, Murat A. Erdogdu.
  To appear in ICLR 2025, International Conference on Learning Representations.

• Learning Multi-Index Models with Neural Networks via Mean-Field Langevin Dynamics.
  Alireza Mousavi-Hosseini, Denny Wu, Murat A. Erdogdu.
  To appear in ICLR 2025, International Conference on Learning Representations.

• Mean-Field Langevin Dynamics for Signed Measures via a Bilevel Approach.
  Guillaume Wang*, Alireza Mousavi-Hosseini*, Lénaïc Chizat.
  NeurIPS 2024, Advances in Neural Information Processing Systems. (Spotlight)

• A Separation in Heavy-Tailed Sampling: Gaussian vs. Stable Oracles for Proximal Samplers.
  Ye He, Alireza Mousavi-Hosseini, Krishnakumar Balasubramanian, Murat A. Erdogdu.
  NeurIPS 2024, Advances in Neural Information Processing Systems.

• Gradient-Based Feature Learning under Structured Data.
  Alireza Mousavi-Hosseini, Denny Wu, Taiji Suzuki, Murat A. Erdogdu.
  NeurIPS 2023, Advances in Neural Information Processing Systems.

• Towards a Complete Analysis of Langevin Monte Carlo: Beyond Poincaré Inequality.
  Alireza Mousavi-Hosseini*, Tyler Farghly*, Ye He, Krishnakumar Balasubramanian, Murat A. Erdogdu.
  COLT 2023, Annual Conference on Learning Theory.

• Neural Networks Efficiently Learn Low-Dimensional Representations with SGD.
  Alireza Mousavi-Hosseini, Sejun Park, Manuela Girotti, Ioannis Mitliagkas, Murat A. Erdogdu.
  ICLR 2023, International Conference on Learning Representations. (Spotlight)

* Equal Contributions

Miscellaneous

• I am fascinated by the performing arts, old and modern theatres, and opera houses.

• UofT has an amazing theatre: Hart House Theatre