In recent years, tasks of machine learning ranging from image processing &
audio/video analysis to natural language understanding have been transformed by
deep learning. The data content in all these scenarios are expressed via
Euclidean space. However, a considerable amount of application data is
structured in non-Euclidean space and is expressed as graphs, e.g. dealing with
complicated interactions & object interdependencies. Modelling physical
systems, learning molecular signatures, identifying protein interactions and
predicting diseases involve utilising a model that can adapt from graph data.
Graph neural networks (GNNs), specified as artificial-neural models, employ
message transmission between graph nodes to represent graph dependencies and
are primarily used in the non-Euclidean domain. Variants of GNN like Graph
Recurrent Networks (GRN), Graph Auto Encoder (GAE), Graph Convolution Networks
(GCN), Graph Adversarial Methods & Graph Reinforcement learning have exhibited
breakthrough productivity on a wide range of tasks, especially in the field of
bioinformatics, in recent years as a result of the rapid collection of
biological network data. Apart from presenting all existing GNN models,
mathematical analysis and comparison of the variants of all types of GNN have
been highlighted in this survey. Graph neural networks are investigated for
their potential real-world applications in various fields, focusing on
Bioinformatics. Furthermore, resources for evaluating graph neural network
models and accessing open-source code & benchmark data sets are included.
Ultimately, we provide some (seven) proposals for future research in this
rapidly evolving domain. GNNs have the potential to be an excellent tool for
solving a wide range of biological challenges in bioinformatics research, as
they are best represented as connected complex graphs.

深度学习在图神经网络中的非欧几里得空间的生物信息学研究中具有潜在的优势，能够解决与生物相关的复杂图形问题。

深度图神经网络的系统综述：挑战、分类、架构、应用与生物信息学的潜在用途

A Systematic Review of Deep Graph Neural Networks: Challenges,  Classification, Architectures, Applications & Potential Utility in  Bioinformatics

The existing sonar image classification methods based on deep learning are
often analyzed in Euclidean space, only considering the local image features.
For this reason, this paper presents a sonar classification method based on
improved Graph Attention Network (GAT), namely SI-GAT, which is applicable to
multiple types imaging sonar. This method quantifies the correlation
relationship between nodes based on the joint calculation of color proximity
and spatial proximity that represent the sonar characteristics in non-Euclidean
space, then the KNN (K-Nearest Neighbor) algorithm is used to determine the
neighborhood range and adjacency matrix in the graph attention mechanism, which
are jointly considered with the attention coefficient matrix to construct the
key part of the SI-GAT. This SI-GAT is superior to several CNN (Convolutional
Neural Network) methods based on Euclidean space through validation of real
data.

本文提出了一种基于改进图形注意力网络 (SI-GAT) 的声纳分类方法，该方法可以在多种类型成像声纳中应用，并通过实际数据验证优于几种以欧几里得空间为基础的卷积神经网络方法。