Graph representation learning resurges as a trending research subject owing to the widespread use of deep learning for Euclidean data, which inspire various creative designs of neural networks in the non-Euclidean domain, particularly graphs. With the success of these graph neural networks (GNN) in the static setting, we approach further practical scenarios where the graph dynamically evolves. Existing approaches typically resort to node embeddings and use a recurrent neural network (RNN, broadly speaking) to regulate the embeddings and learn the temporal dynamics. These methods require the knowledge of a node in the full time span (including both training and testing) and are less applicable to the frequent change of the node set. In some extreme scenarios, the node sets at different time steps may completely differ. To resolve this challenge, we propose EvolveGCN, which adapts the graph convolutional network (GCN) model along the temporal dimension without resorting to node embeddings. The proposed approach captures the dynamism of the graph sequence through using an RNN to evolve the GCN parameters. Two architectures are considered for the parameter evolution. We evaluate the proposed approach on tasks including link prediction, edge classification, and node classification. The experimental results indicate a generally higher performance of EvolveGCN compared with related approaches. The code is available at https://github.com/IBM/EvolveGCN.

A sol-gel method was applied for the development of highly permeable hydrogen separation membranes using bis(triethoxysilyl)ethane (BTESE) as a silica precursor. Hybrid silica membranes showed quite high hydrogen permeance (1 x 10(-5) mol m(-2) s(-1) Pa(-1)) with a high H(2)-to-SF(6) selectivity of 1000 because of loose organic-inorganic silica networks. Hybrid silica membranes were found to show high hydrothermal stability due to the presence of Si-C-C-Si bonds in silica networks.

No abstract

in Wiley InterScience (www.interscience.wiley.com).DDR-type zeolite membranes were prepared by the secondary growth method on porous a-alumina disk, followed by on-stream counter diffusion chemical vapor deposition modification to eliminate the intercrystalline micropores. Single gas permeation of He, H 2 , CO 2 , and CO through this zeolite membrane before and after CVD modification was measured in 25-5008C. Intracrystalline diffusivities for these four gases in DDR-type zeolite were obtained from the permeation data above 3008C to examine the effects of the size and molecular weight of permeating gases on diffusion and permeation rate for this zeolite membrane. For the unmodified DDR-type zeolite membrane with presence of a small amount intercrystalline micropores the diffusivity (or permeance) with a low activation energy depends on both the size and molecular weight of permeating gases. For the CVD-modified DDR-type zeolite membrane with intercrystalline micropores eliminated, the activation energy for diffusion and diffusivity increases with increasing molecular size of the permeating gases.

No abstract