Authors: Mathias Niepert, Mohamed Ahmed, Konstantin Kutzkov
Partners involved: NEC
Proceeding: Proceedings of the 33rd International Conference on Machine Learning
Abstract: Numerous important problems can be framed as learning from graph data. We propose a framework for learning convolutional neural networks for arbitrary graphs. These graphs may be undirected, directed, and with both discrete and continuous node and edge attributes. Analogous to image-based convolutional networks that operate on locally connected regions of the input, we present a general approach to extracting locally connected regions from graphs. Using established benchmark data sets, we demonstrate that the learned feature representations are competitive with state of the art graph kernels and that their computation is highly efficient.
- We present a method that allow the analysis of graph data (as the Web or the relations among advertisers and publishers) using deep learning.
- The method presented needs less hyper-parameters and is faster than the state of the art methods.
Article available here