No abstract

Traditional Graph Neural Networks (GNNs) rely on message passing, which amounts to permutation-invariant local aggregation of neighbour features. Such a process is isotropic and there is no notion of 'direction' on the graph. We present a new GNN architecture called Graph Anisotropic Diffusion. Our model alternates between linear diffusion, for which a closed-form solution is available, and local anisotropic filters to obtain efficient multi-hop anisotropic kernels. We test our model on two common molecular property prediction benchmarks (ZINC and QM9) and show its competitive performance.

Original Article Surgical Operations Ahmed AA Elhag of the total number of cholecystectomies. General anaesthesia was used in 57.6% of operations. Urological operations represented 13.1% and paediatric surgery 11.7% of the total number of operations. Most of the patients (97%, n=846) were discharged home in good condition. Conclusion A variety of surgical operations were performed with a good outcome. Appendicectomy and hernia repair were the most common operations.

No abstract