Unsupervised Neighborhood Propagation Kernel Layers for Semi-supervised Node Classification


  • Sonny Achten KU Leuven
  • Francesco Tonin KU Leuven
  • Panagiotis Patrinos KU Leuven
  • Johan A.K. Suykens KU Leuven




ML: Kernel Methods, ML: Graph-based Machine Learning, ML: Semi-Supervised Learning, ML: Unsupervised & Self-Supervised Learning, ML: Optimization, ML: Learning with Manifolds, ML: Clustering


We present a deep Graph Convolutional Kernel Machine (GCKM) for semi-supervised node classification in graphs. The method is built of two main types of blocks: (i) We introduce unsupervised kernel machine layers propagating the node features in a one-hop neighborhood, using implicit node feature mappings. (ii) We specify a semi-supervised classification kernel machine through the lens of the Fenchel-Young inequality. We derive an effective initialization scheme and efficient end-to-end training algorithm in the dual variables for the full architecture. The main idea underlying GCKM is that, because of the unsupervised core, the final model can achieve higher performance in semi-supervised node classification when few labels are available for training. Experimental results demonstrate the effectiveness of the proposed framework.



How to Cite

Achten, S., Tonin, F., Patrinos, P., & Suykens, J. A. (2024). Unsupervised Neighborhood Propagation Kernel Layers for Semi-supervised Node Classification. Proceedings of the AAAI Conference on Artificial Intelligence, 38(10), 10766-10774. https://doi.org/10.1609/aaai.v38i10.28949



AAAI Technical Track on Machine Learning I