DFNets: Spectral CNNs for Graphs with Feedback-Looped Filters
This addresses graph-based classification problems for researchers and practitioners, offering improved accuracy with theoretical guarantees, though it appears incremental as it builds on existing spectral CNN methods.
The authors tackled graph-structured data classification by proposing DFNets, a spectral CNN with feedback-looped filters, which achieved state-of-the-art performance on semi-supervised document and entity classification benchmarks.
We propose a novel spectral convolutional neural network (CNN) model on graph structured data, namely Distributed Feedback-Looped Networks (DFNets). This model is incorporated with a robust class of spectral graph filters, called feedback-looped filters, to provide better localization on vertices, while still attaining fast convergence and linear memory requirements. Theoretically, feedback-looped filters can guarantee convergence w.r.t. a specified error bound, and be applied universally to any graph without knowing its structure. Furthermore, the propagation rule of this model can diversify features from the preceding layers to produce strong gradient flows. We have evaluated our model using two benchmark tasks: semi-supervised document classification on citation networks and semi-supervised entity classification on a knowledge graph. The experimental results show that our model considerably outperforms the state-of-the-art methods in both benchmark tasks over all datasets.