Eitan Levin

Publications

1. Any-Dimensional Polynomial Optimization via de Finetti Theorems
   with Venkat Chandrasekaran. Submitted. 2025.
   [code] [slides]

2. Poset-Markov Channels: Capacity via Group Symmetry
   with Eray U. Atay, Venkat Chandrasekaran, and Victoria Kostina. Submitted. 2025.

3. On Transferring Transferability: Towards a Theory for Size Generalization
   with Yuxin Ma, Mateo Díaz, and Soledad Villar. NeurIPS. 2025. (Spotlight)
   [code]

4. Free descriptions of convex sets
   with Venkat Chandrasekaran. Submitted. 2024.
   [code] [slides]

5. Any-dimensional equivariant neural networks
   with Mateo Díaz. AISTATS. 2024.
   [journal] [code] [poster]

6. The effect of smooth parametrizations on nonconvex optimization landscapes
   with Joe Kileel and Nicolas Boumal. Mathematical Programming. 2024. (INFORMS optimization society student paper prize)
   [journal] [slides]

7. Finding stationary points on bounded-rank matrices: A geometric hurdle and a smooth remedy
   with Joe Kileel and Nicolas Boumal. Mathematical Programming. 2023.
   [journal]

8. Towards optimization on varieties
   Undergraduate senior thesis. Princeton University. 2020.
   [DataSpace]

9. A note on Douglas-Rachford, gradients, and phase retrieval
   with Tamir Bendory. arXiv:1911.13179. 2019.

10. Multi-target detection with application to cryo-electron microscopy
    with Tamir Bendory, Nicolas Boumal, William Leeb, and Amit Singer. Inverse Problems. 2019.
    [journal] [code]

11. Toward single particle reconstruction without particle picking: Breaking the detection limit
    with Tamir Bendory, Nicolas Boumal, William Leeb, and Amit Singer. SIAM Journal on Imaging Sciences. 2023.
    [journal] [code]

12. 3D ab initio modeling in cryo-EM by autocorrelation analysis
    with Tamir Bendory, Nicolas Boumal, Joe Kileel, and Amit Singer. 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018). 2018. (Best student paper)
    [journal] [code]

13. Direct reconstruction of two-dimensional currents in thin films from magnetic-field measurements
    with Alexander Meltzer and Eli Zeldov. Physical Review Applied. 2017.
    [journal] [code]

14. Stopping criterion for iterative regularization of large-scale ill-posed problems using the Picard parameter
    with Alexander Meltzer. arXiv:1707.04200. 2017.
    [code]

15. Estimation of the regularization parameter in linear discrete ill-posed problems using the Picard parameter
    with Alexander Meltzer. SIAM Journal on Scientific Computing. 2017.
    [journal] [code]