Improving Graph Convolutional Networks with Non-Parametric Activation Functions
This work addresses a specific bottleneck in graph neural networks for domains like citation networks or knowledge graphs, though it is incremental as it builds on existing kernel activation functions.
The paper tackles the problem of limited nonlinearity in graph convolutional networks (GCNs) by introducing non-parametric activation functions that adapt from training data, resulting in significant improvements over baseline GCNs without increasing model depth or size.
Graph neural networks (GNNs) are a class of neural networks that allow to efficiently perform inference on data that is associated to a graph structure, such as, e.g., citation networks or knowledge graphs. While several variants of GNNs have been proposed, they only consider simple nonlinear activation functions in their layers, such as rectifiers or squashing functions. In this paper, we investigate the use of graph convolutional networks (GCNs) when combined with more complex activation functions, able to adapt from the training data. More specifically, we extend the recently proposed kernel activation function, a non-parametric model which can be implemented easily, can be regularized with standard $\ell_p$-norms techniques, and is smooth over its entire domain. Our experimental evaluation shows that the proposed architecture can significantly improve over its baseline, while similar improvements cannot be obtained by simply increasing the depth or size of the original GCN.