ABSTRACT
Link prediction in complex networks aims to mine hidden or to-be-generated links between network nodes, which plays a significant role in fields such as the cold start of recommendation systems, knowledge graph completion and biomedical experiments. The existing link prediction models based on graph neural networks, such as graph convolution neural networks, often only learn the low-frequency information reflecting the common characteristics of nodes while ignoring the high-frequency information reflecting the differences between nodes when learning node representation, which makes the corresponding link prediction models show over smoothness and poor performance. Focusing on links in complex networks, this paper proposes an edge convolutional graph neural network EdgeConvHiF that fuses high-frequency node information to achieve the representation learning of links so that link prediction can be realized by implementing the classification of links. EdgeConvHiF can also be employed as a baseline, and extensive experiments on real-world benchmarks validate that EdgeConvHiF not only has high stability but also has more advantages than the existing representative baselines.
ABSTRACT
A magnet system is used in the SPERF to create the magnetic field configuration for simulating the space plasma environment. In this paper, the parameters of the system are designed to achieve the target fields needed by the scaling laws, and the electromagnetic analysis has been performed to validate the results. A procedure to obtain the parameters is proposed based on the investigation into the physical and technological constraints. The vacuum magnetic fields for studying the 3D magnetic reconnection at the magnetopause, Earth's magnetosphere, and 3D magnetic reconnection driven by a plasma gun are computed. In addition, the engineering complexity is reviewed in brief. This research is crucial to the construction of the SPERF, and it is valuable to designing the magnets applied in other fields.