I am an incoming assistant professor of computer science at Oberlin College. I finished my PhD in 2026 at UC-San Diego and was advised by Yusu Wang. Before that, I did my undergraduate degree in computer science and math at Carleton College. Currently, I am interested in research related to neural algorithmic reasoning, geometric deep learning, computational geometry, geospatial analysis, and optimal transport. If you are an undergraduate at Oberlin interested in doing research on any of these topics, feel free to contact me!

Publications

Neural Algorithmic Reasoning for Graph Saddle-Point Problems (S. Chen, C. Clougherty, J. He, G. Mishne, C. Holtz, in submission 2026)

Efficient Neural Approximations of Geometric Optimization Problems (S. Chen, O. Ciolli, A. Sidiropolous, Y. Wang, accepted to NeurIPS 2025)

De-Coupled NeuroGF for Shortest Path Distance Approximations on Large Terrains (S. Chen, P. K. Agarwal, Y. Wang, accepted to ICML 2025)

Graph neural networks extrapolate out-of-distribution for shortest paths (R. R. Nerem, S. Chen, S. Dasgupta, Y. Wang, preprint 2025)

Neural approximations of Wasserstein distance via a universal architecture for symmetric and factor-wise group invariant functions (S. Chen, Y. Wang, accepted to NeurIPS 2023)

Learning Ultrametric Trees for Optimal Transport Regression (S. Chen, P. Tabaghi, Y. Wang, accepted to AAAI 2024)

The Weisfeiler-Lehman Distance: Reinterpretation and Connetion GNNs (S. Chen, S. Lim, F. Mémoli, Z. Wan, Y. Wang, ICML workshop: Topology, Algebra, and Geometry in Machine Learning (2023))

Weisfeiler-Lehman meets Gromov-Wasserstein (S. Chen, S. Lim, F. Mémoli, Z. Wan, Y. Wang, accepted to ICML 2022)

Approximation algorithms for 1-Wasserstein distance between persistence diagrams (S. Chen, Y. Wang, accepted to SEA 2021, extended version in Computational Geometry, 2025)