Adaptive Kernel Graph Neural Network

Ju, Mingxuan; Hou, Shifu; Yu, Fan; Zhao, Jianan; Ye, Yanfang; Zhao, Ling

doi:10.1609/aaai.v36i6.20664

Cited by 11 publications

(1 citation statement)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Zhang et al [66] propose a sort pooling to generate the graph-level representation by sorting the final node representations. Ju et al [34] present a layer-wise readout by extending the node information aggregated from the last layer of GNNs to all layers. However, none of the existing readout functions leverages the properties of brain networks that nodes in the same functional modules tend to have similar behaviors and clustered representations, as shown in Figure 1(a).…”

Section: Orthonormal Clustering Readout (Ocread)mentioning

confidence: 99%

Brain Network Transformer

Kan¹,

Dai²,

Cui³

et al. 2022

Preprint

View full text Add to dashboard Cite

Human brains are commonly modeled as networks of Regions of Interest (ROIs) and their connections for the understanding of brain functions and mental disorders. Recently, Transformer-based models have been studied over different types of data, including graphs, shown to bring performance gains widely. In this work, we study Transformer-based models for brain network analysis. Driven by the unique properties of data, we model brain networks as graphs with nodes of fixed size and order, which allows us to (1) use connection profiles as node features to provide natural and low-cost positional information and (2) learn pairwise connection strengths among ROIs with efficient attention weights across individuals that are predictive towards downstream analysis tasks. Moreover, we propose an ORTHONORMAL CLUSTERING READOUT operation based on selfsupervised soft clustering and orthonormal projection. This design accounts for the underlying functional modules that determine similar behaviors among groups of ROIs, leading to distinguishable cluster-aware node embeddings and informative graph embeddings. Finally, we re-standardize the evaluation pipeline on the only one publicly available large-scale brain network dataset of ABIDE, to enable meaningful comparison of different models. Experiment results show clear improvements of our proposed BRAIN NETWORK TRANSFORMER on both the public ABIDE and our restricted ABCD datasets. The implementation is available at https://github.com/Wayfear/BrainNetworkTransformer. Nowadays Transformer-based models have led a tremendous success in various downstream tasks across fields including natural language processing [56, 17] and computer vision [20, 10, 55]. Recent efforts have also emerged to apply Transformer-based designs to graph representation learning. GAT [57] firstly adapts the attention mechanism to graph neural networks (GNNs) but only considers the local structures of neighboring nodes. Graph Transformer [21] injects edge information into the attention mechanism and leverages the eigenvectors of each node as positional embeddings. SAN [40] further enhances the positional embeddings by considering both eigenvalues and eigenvectors and improves the attention mechanism by extending the attention from local to global structures.36th Conference on Neural Information Processing Systems (NeurIPS 2022).

show abstract

Section: Orthonormal Clustering Readout (Ocread)mentioning

confidence: 99%

Brain Network Transformer

Kan¹,

Dai²,

Cui³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

ESS MS-G3D: extension and supplement shift MS-G3D network for the assessment of severe mental retardation

Liu,

Cai,

Liu

et al. 2023

Complex Intell. Syst.

View full text Add to dashboard Cite

Automated mental retardation (MR) assessment is potential for improving the diagnostic efficiency and objectivity in clinical practice. Based on the researches on abnormal behavior characteristics of patients with MR, we propose an extension and supplement shift multi-scale G3D (ESS MS-G3D) network for video-based assessment of MR. Specifically, all videos are collected from clinical diagnostic scenarios and the skeleton sequence of human body is extracted from videos through an advanced pose estimation model. To solve the shortcomings of existing behavior characteristic learning methods, we present: (1) three G3D styles, enable the network to have different input forms; (2) two G3D graphs and two extension graphs, redefine and extend the graph structure of spatial–temporal nodes; (3) two learnable parameters, realize adaptive adjustment of graph structure; (4) a shift layer, enable the network to learn global features. Finally, we construct a three-branch model ESS MS-STGC, which can capture the discriminative spatial–temporal features and explore the co-occurrence relationship between spatial and temporal domains. Experiments in clinical video data set show that our proposed model has good performance in MR assessment and is superior to the existing vision-based methods. In two-classification task, our model with joint stream achieves the highest accuracy of $$94.63\%$$ 94.63 % in validation set and $$89.13\%$$ 89.13 % in test set. The results are further improved to $$96.52\%$$ 96.52 % and $$93.22\%$$ 93.22 % , respectively, by utilizing multi-stream fusion strategy. In four-classification task, our model obtains Top1 accuracy of $$78.84\%$$ 78.84 % and Top2 accuracy of $$91.34\%$$ 91.34 % in test set. The proposed method provides a new idea for clinical mental retardation assessment.

show abstract

Interpretable decision-making model with uncertain weights for sustainable digital economy

Wang

2024

Advanced Engineering Informatics

View full text Add to dashboard Cite

Adaptive Kernel Graph Neural Network

Cited by 11 publications

References 17 publications

Brain Network Transformer

Brain Network Transformer

ESS MS-G3D: extension and supplement shift MS-G3D network for the assessment of severe mental retardation

Interpretable decision-making model with uncertain weights for sustainable digital economy

Contact Info

Product

Resources

About