Contextual Stochastic Block Model (CSBM)

The Contextual Stochastic Block Model (CSBM) is a benchmark for evaluating the effects of graph convolutions on node representations in random graphs, specifically focusing on the balance between mixing and denoising effects that contribute to oversmoothing in Graph Neural Networks.

Papers using Contextual Stochastic Block Model (CSBM) (1)

A Non-Asymptotic Analysis of Oversmoothing in Graph Neural Networks2022 · 9 cites