Publications

Levi Rauchwerger, Stefanie Jegelka, and Ron Levie. Neural Networks With Dense Weights Are Not Universal Approximators, NeurIPS 2026. Also presented at the GFM Workshop, ICML 2026.

Levi Rauchwerger and Ron Levie. A Note on Graphon-Signal Analysis of Graph Neural Networks, Preprint. 2025.

Levi Rauchwerger, Stefanie Jegelka, and Ron Levie. Generalization, Expressivity, and Universality of Graph Neural Networks on Attributed Graphs, ICLR. 2025.

Ink and watercolor illustration of wave patterns flowing into branching language-like structures