An End-To-End Hyperbolic Deep Graph Convolutional Neural Network Framework

Graph Convolutional Neural Networks(GCNs)have been widely used in various fields due to their powerful capabilities in processing graph-structured data.However,GCNs encounter significant challenges when applied to scale-free graphs with power-law distributions,resulting in substantial distortions.Moreover,most of the existing GCN models are shallow structures,which restricts their ability to capture dependencies among distant nodes and more refined high-order node features in scale-free graphs with hierarchical structures.To more broadly and precisely apply GCNs to real-world graphs exhibiting scale-free or hierarchical structures and utilize multi-level aggregation of GCNs for capturing high-level information in local representations,we propose the Hyperbolic Deep Graph Convolutional Neural Network(HDGCNN),an end-to-end deep graph representation learning framework that can map scale-free graphs from Euclidean space to hyperbolic space.In HDGCNN,we define the fundamental operations of deep graph convolutional neural networks in hyperbolic space.Additionally,we introduce a hyperbolic feature transformation method based on identity mapping and a dense connection scheme based on a novel non-local message passing framework.In addition,we present a neighborhood aggregation method that combines initial structural featureswith hyperbolic attention coefficients.Through the above methods,HDGCNN effectively leverages both the structural features and node features of graph data,enabling enhanced exploration of non-local structural features and more refined node features in scale-free or hierarchical graphs.Experimental results demonstrate that HDGCNN achieves remarkable performance improvements over state-ofthe-art GCNs in node classification and link prediction tasks,even when utilizing low-dimensional embedding representations.Furthermore,when compared to shallow hyperbolic graph convolutional neural network models,HDGCNN exhibits notable advantages and performance enhancements.

High-precision temperature retrieval algorithm and verification for mesospheric airglow spectrum photometer

Photometric technology,characterized by its compact structure and relatively high stability,finds wide application in measuring airglow spectra.This instrumentation is anticipated to assume a pivotal role as the primary equipment for extensive network observations of middle and upper atmospheric temperatures in China,thereby providing crucial support for space environmental monitoring and atmospheric dynamic research.Nevertheless,susceptibility to various factors such as instrument inconsistency,variability in observation conditions,and alterations in the background atmospheric environment across different stations poses a challenge,potentially resulting in data inconsistencies in network observations.In response to these challenges,we propose a multiple-parameter iterative inversion(MPII)algorithm for temperature retrieval based on a mesospheric airglow spectrum photometer(MASP)developed by our research group.This algorithm accurately identifies the center of the image circle,corrects image distortion,and thereby obtains an accurate synthetic spectrum reflective of actual observations.It encompasses five adjustable parameters:sky background light,atmospheric temperature,filter temperature,optical system focal length,and degree of synthetic spectrum modulation.Compared to traditional methods,significant enhancements in the accuracy of the inverted temperature are achieved.To validate the effectiveness of the MPII algorithm,we conducted combined active and passive remote sensing synchronous measurements using MASP in conjunction with a sodium fluorescence Doppler lidar developed by the National Space Science Center.By utilizing the lidar temperature as a reference,atmospheric background radiation is mitigated from the MASP data,and the temperature is inverted using the MPII algorithm.Comparative analysis with the traditional method reveals that temperatures calculated by the MPII algorithm exhibit better consistency than those observed by the lidar.