My name is Szymon Gustav Snoeck and I am a fourth-year applied mathematics major at Columbia University, minoring in computer science. My main research interests are in algorithms and machine learning for high-dimensional data. However, through guided research projects and graduate classes, I have also done research in nearest-neighbor search, learning theory, matching theory, dimension-reduction, and derandomization. Some of the work I have produced can be found below.

Update: I am happy to share that I was selected as a finalist for the CRA Outstanding Undergraduate Researcher Award, and my two papers have been accepted at ALT 2026 and ICLR 2026.

• t-SNE Exaggerates Clusters, Provably.
  Noah Bergam, Szymon Snoeck, Nakul Verma.
  International Conference on Learning Representations (ICLR), 2026. arxiv
• Compressibility Barriers to Neighborhood-Preserving Data Visualizations.
  Szymon Snoeck, Noah Bergam, Nakul Verma.
  Algorithmic Learning Theory (ALT), 2026. arxiv

• A Uniform Convergence Result for Learning Text Data.
  Szymon Snoeck.
  Manuscript, 2025. pdf
• The Negative Inter-Dependencies of the Multivariate Hypergeometric Distribution.
  Szymon Snoeck.
  Manuscript, 2025. pdf
• The Difficulty of Approximating NSW in Online Matching.
  Szymon Snoeck, Christopher En, Yuri Faenza.
  Manuscript, 2024. pdf
• Deterministic Approximate Counting F2 Polynomials Via Correlation-based Fourier Bounds.
  Szymon Snoeck, Sam Wang.
  Manuscript, 2024. pdf