I am a PhD candidate in Applied Mathematics at Tel Aviv University, jointly supervised by Dr. Inbar Seroussi and Dr. Dmitry Batenkov.

My research focuses on mathematical models of neural networks in high-dimensional settings. Previously, I completed my MSc at Tel Aviv University, where I studied physics-informed neural networks for atmospheric radiative transfer.

Research interests

  • Learning theory, transformers, and in-context learning
  • Learning collective dynamics and probabilistic forecasting
  • Physics-informed neural networks

Publications

  1. 2026

    Minimax Rates for Learning Pairwise Interactions in Attention-Style Models

    S. Zucker, X. Wang, F. Lu, and I. Seroussi

    International Conference on Learning Representations (ICLR), 2026

    Paper on OpenReview
  2. 2025

    Physics-informed neural networks for modeling atmospheric radiative transfer

    S. Zucker, D. Batenkov, and M. S. Rozenhaimer

    Journal of Quantitative Spectroscopy and Radiative Transfer, 331, 109253

    Published article

Preprints

Education

2023–present
PhD, Applied MathematicsTel Aviv University
2021–2023
MSc, Applied MathematicsTel Aviv University · Summa cum laude

Thesis: Physics-Informed Neural Networks for Modeling Atmospheric Radiative Transfer.

2013–2016
BSc, Applied MathematicsBar-Ilan University

Program for gifted youth in mathematics.

Teaching

Teaching assistant at Tel Aviv University for undergraduate engineering courses, 2021–2022 and 2024–2026.

  • Harmonic Analysis
  • Ordinary Differential Equations

Talks & posters

  • July 2026 · Student talkIsraeli Mathematical Union (IMU) Student Talks

    Minimax Rates for Learning Pairwise Interactions in Attention-Style Models

  • 2026 · Student spotlight talkNUS Summer School on Mathematical Aspects of Data Science

    Minimax Rates for Learning Pairwise Interactions in Attention-Style Models

  • 2025 · PosterPrinceton Machine Learning Theory Summer School

    Minimax Rates for Transformer Attention via Interacting Particle Systems

  • 2024 · Contributed talkSIAM Conference on Mathematics of Planet Earth (MPE24)

    Physics-Informed Neural Networks for Modeling Atmospheric Radiative Transfer

  • 2022 · Contributed talkInternational Radiation Symposium (IRS)

    Utilization of Convolutional Neural Networks and Physics-Informed Neural Networks to Retrieve Cloud and Aerosol Properties from Polarimetric Observations

Awards & distinctions

  • 2026
    KLA Scholarship for Outstanding Doctoral Students
  • 2026
    Honorable Mention, Israel Mathematical Union student talks
  • 2026, 2023
    Tel Aviv University Award Certificate for Outstanding Math Students