A Brief Review of Receptive Fields in Graph Convolutional Networks

Quan, Pei; Shi, Yong; Lei, Minglong; Leng, Jiaxu; Zhang, Tianlin; Niu, Lingfeng

doi:10.1145/3358695.3360934

Cited by 11 publications

(3 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[31] introduced GNNs based on small graph and giant graph respectively. [32], [33] focused on reviewing related works in a specific branch of GNNs, i.e., graph convolutional network (GCN). However, they seldom introduce GNNs works related with traffic scenarios.…”

Section: Related Workmentioning

confidence: 99%

How to Build a Graph-Based Deep Learning Architecture in Traffic Domain: A Survey

Ye,

Zhao,

et al. 2020

Preprint

View full text Add to dashboard Cite

In recent years, various deep learning architectures have been proposed to solve complex challenges (e.g., spatial dependency, temporal dependency) in traffic domain, which have achieved satisfactory performance. These architectures are composed of multiple deep learning techniques in order to tackle various challenges in traffic data. Traditionally, convolution neural networks (CNNs) are utilized to model spatial dependency by decomposing the traffic network as grids. However, many traffic networks are graph-structured in nature. In order to utilize such spatial information fully, it's more appropriate to formulate traffic networks as graphs mathematically. Recently, various novel deep learning techniques have been developed to process graph data, called graph neural networks (GNNs). More and more works combine GNNs with other deep learning techniques to construct an architecture dealing with various challenges in a complex traffic task, where GNNs are responsible for extracting spatial correlations in traffic network. These graphbased architectures have achieved state-of-the-art performance. To provide a comprehensive and clear picture of such emerging trend, this survey carefully examines various graph-based deep learning architectures in many traffic applications. We first give guidelines to formulate a traffic problem based on graph and construct graphs from various traffic data. Then we decompose these graph-based architectures to discuss their shared deep learning techniques, clarifying the utilization of each technique in traffic tasks. What's more, we summarize common traffic challenges and the corresponding graph-based deep learning solutions to each challenge. Finally, we provide benchmark datasets, open source codes and future research directions in this rapidly growing field.

show abstract

Section: Related Workmentioning

confidence: 99%

How to Build a Graph-Based Deep Learning Architecture in Traffic Domain: A Survey

Ye,

Zhao,

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Conjugate graphs have also been applied in the field of shape matching, where a novel formalism based on the conjugate product graph has been proposed to find continuous and non-rigid matchings between 2D contours and 3D meshes [4]. Additionally, the conjugate graph has been used to study the isomorphism between nonabelian groups and to compare the conjugate graphs of different groups [5]. Overall, conjugate graphs have been extensively studied in various contexts, including group theory, graph theory, and shape matching.…”

Section: Introductionmentioning

confidence: 99%

Machine Learning Driven Exploration of Energies and Generalization of Topological Indices for the Fuzzy Conjugate Graph of Dihedral Group

Umar Mirza,

Anjum,

Fahad Alrasheedi

et al. 2024

IEEE Access

View full text Add to dashboard Cite

This study introduces a ground-breaking approach to analyzing dihedral groups through the lens of fuzzy graph theory, significantly enhancing computational efficiency in group theory. By extending the fuzzy topological indices to polynomial forms, this research drastically reduces the calculation time for these indices from hours to mere seconds. A notable feature of this work is the innovative use of polynomial regression, a machine learning technique, to generate polynomials for adjacency and degree matrices within the fuzzy conjugate graph of a dihedral group. This method not only simplifies calculations but also incorporates an error analysis component, ensuring accuracy and reliability. The integration of fuzzy graph theory with polynomial regression in this context is a pioneering step, offering valuable insights into the structural attributes of dihedral groups. This research goes beyond traditional methods, highlighting the effectiveness of machine learning in deciphering complex patterns in group theory. The findings and techniques presented hold great promise for future applications in graph theory and group theory, offering a novel perspective for understanding and analyzing intricate graph structures. This study stands as a significant contribution to the field, potentially revolutionizing the way complex mathematical problems are approached and solved.

show abstract

“…The literature examines Graph Neural Networks (GNNs) in-depth, covering general knowledge, network structures, potential gaps, and perspectives through various surveys [1,2,12,15,18,19,[36][37][38][39][40][41][42][43][44][45][46][47]. In addition, the existing literature includes surveys and reviews focusing on specific applications [48][49][50][51].…”

Section: Introductionmentioning

confidence: 99%

A Survey of Computationally Efficient Graph Neural Networks for Reconfigurable Systems

Kose,

Nunez-Yanez,

Piechocki

et al. 2024

Preprint

View full text Add to dashboard Cite

Graph Neural Networks (GNNs) are powerful models capable of managing intricate connections in non-Euclidean data, such as social networks, physical systems, chemical structures, and communication networks. Despite their effectiveness, the large-scale and complex nature of graph data demands substantial computational resources and high performance during both training and inference stages, presenting significant challenges, particularly in the context of embedded systems. Recent studies on GNNs have investigated both software and hardware solutions to enhance computational efficiency. Earlier studies on Deep Neural Networks (DNNs) have indicated that methods like reconfigurable hardware and quantization are beneficial in addressing these issues. Unlike DNN research, studies on efficient computational methods for GNNs are less developed and require more exploration. This survey reviews the latest developments in quantization and FPGA-based acceleration for GNNs, showcasing the capabilities of reconfigurable systems (often FPGAs) to offer customized solutions in environments marked by significant sparsity and the necessity for dynamic load management. It also emphasizes the role of quantization in reducing both computational and memory demands through the use of fixed-point arithmetic and streamlined vector formats. This paper concentrates on low-power, resource-limited devices over general hardware accelerators and reviews research applicable to embedded systems. Additionally, it provides a detailed discussion of potential research gaps, foundational knowledge, obstacles, and prospective future directions.

show abstract

A Brief Review of Receptive Fields in Graph Convolutional Networks

Cited by 11 publications

References 5 publications

How to Build a Graph-Based Deep Learning Architecture in Traffic Domain: A Survey

How to Build a Graph-Based Deep Learning Architecture in Traffic Domain: A Survey

Machine Learning Driven Exploration of Energies and Generalization of Topological Indices for the Fuzzy Conjugate Graph of Dihedral Group

A Survey of Computationally Efficient Graph Neural Networks for Reconfigurable Systems

Contact Info

Product

Resources

About