Over-smoothing Effect of Graph Convolutional Networks
This addresses a key limitation for researchers and practitioners using deep graph neural networks, though it is incremental as it builds on prior work.
The paper tackled the over-smoothing problem in Graph Convolutional Networks by analyzing its mechanism and deriving an upper bound for its occurrence, which explains the feasibility of existing algorithms to mitigate it.
Over-smoothing is a severe problem which limits the depth of Graph Convolutional Networks. This article gives a comprehensive analysis of the mechanism behind Graph Convolutional Networks and the over-smoothing effect. The article proposes an upper bound for the occurrence of over-smoothing, which offers insight into the key factors behind over-smoothing. The results presented in this article successfully explain the feasibility of several algorithms that alleviate over-smoothing.