Hierarchical Pooling in Graph Neural Networks to Enhance Classification Performance in Large Datasets.

Deep learning methods predicated on convolutional neural networks and graph neural networks have enabled significant improvement in node classification and prediction when applied to graph representation with learning node embedding to effectively represent the hierarchical properties of graphs. An interesting approach (DiffPool) utilises a differentiable graph pooling technique which learns 'differentiable soft cluster assignment' for nodes at each layer of a deep graph neural network with nodes mapped on sets of clusters. However, effective control of the learning process is difficult given the inherent complexity in an 'end-to-end' model with the potential for a large number parameters (including the potential for redundant parameters).

Intensified salinity intrusion in coastal aquifers due to groundwater overextraction: a case study in the Mekong Delta, Vietnam.

Groundwater salinization is one of the most severe environmental problems in coastal aquifers worldwide, causing exceeding salinity in groundwater supply systems for many purposes. High salinity concentration in groundwater can be detected several kilometers inland and may result in an increased risk for coastal water supply systems and human health problems. This study investigates the impacts of groundwater pumping practices and regional groundwater flow dynamics on groundwater flow and salinity intrusion in the coastal aquifers of the Vietnamese Mekong Delta using the SEAWAT model-a variable-density groundwater flow and solute transport model.

Upscaling of transport through discrete fracture networks via random walk: A comparison of models.

Many models have been created for upscaled transport modeling in discrete fracture networks (DFNs). Random walk examples of these are the Markov directed random walk (MDRW), Monte Carlo solution of the Boltzmann transport equation (BTE), and the spatial Markov model (SMM). Each model handles the correlation between the random walk steps using different techniques and has successfully reproduced the results of full-resolution transport simulations in DFNs.

Gold nanorod/molybdenum sulfide core-shell nanostructures synthesized by a photo-induced reduction process.

Coupling of plasmonic nanostructures and semiconductors gives promising hybrid nanostructures that can be used in different applications such as photosensing and energy conversion. In this report, we describe an approach for fabricating a new hybrid material by coupling a gold nanorod (Au NR) core and amorphous molybdenum sulfide (MoS) shell. The Au NR/MoS core-shell structure is achieved by exploiting the hot electrons generated in the plasmonic excitation of Au NRs to drive the reduction of [MoS], which is pre-adsorbed on the Au NR surface, producing a thin MoS layer.

Insights into the Electrochemical Polymerization of [Mo S ] Generating Amorphous Molybdenum Sulfide.

Amorphous molybdenum sulfide is an attractive electrode material for Li/Mg batteries and an efficient Pt-free catalyst for the hydrogen evolution reaction in water. By using the electrochemical quartz crystal microbalance (EQCM) analysis, new insights were gained into the electrochemical polymerization of the [Mo S ] cluster, which generates amorphous molybdenum sulfide thin films. In this work, it is shown that, at the anodic potential, a two-electron oxidative elimination of the terminal disulfide ligand within the [Mo S ] cluster induces the polymerization.

Bayesian Chance-Constrained Hydraulic Barrier Design under Geological Structure Uncertainty.

The groundwater community has widely recognized geological structure uncertainty as a major source of model structure uncertainty. Previous studies in aquifer remediation design, however, rarely discuss the impact of geological structure uncertainty. This study combines chance-constrained (CC) programming with Bayesian model averaging (BMA) as a BMA-CC framework to assess the impact of geological structure uncertainty in remediation design.

Identifying and measuring land-use and proximity conflicts: methods and identification.

Abstract: This text aims to present the methodology of study of land-use conflicts performed in recent years by a multidisciplinary team, and to reveal the methods of survey and data collection, as well as the structure of the resulting database. We first define the scope of our study by providing a definition of these conflicts, of their characteristics and motives, of the ways they manifest themselves and of the actors involved (I). We then present the methodology we have used to identify conflicts; it is based on a spatial analysis and the combined use of different data collection methods including surveys conducted by experts, analyses of the regional daily press and of data from the administrative litigation courts (II).