Beyond Localized Graph Neural Networks: An Attributed Motif Regularization Framework

Ashok, S.; Wang, Junting; Krishnan, A.; Sundaram, Hari

doi:10.1109/icdm50108.2020.00056

Cited by 15 publications

(4 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Node properties such as degree, proximity, and attributes, which are seen as local structure information, are often used as the ground truth to fully exploit the unlabeled data [17]. For example, InfoMotif [31] models attribute correlations in motif structures with mutual information maximization to regularize graph neural networks. Meanwhile, global structure information like node pair distance is also harnessed to facilitate representation learning [35].…”

Section: Self-supervised Learningmentioning

confidence: 99%

Self-Supervised Multi-Channel Hypergraph Convolutional Network for Social Recommendation

Yin

et al. 2021

Proceedings of the Web Conference 2021

272

119

View full text Add to dashboard Cite

Social relations are often used to improve recommendation quality when user-item interaction data is sparse in recommender systems. Most existing social recommendation models exploit pairwise relations to mine potential user preferences. However, real-life interactions among users are very complicated and user relations can be high-order. Hypergraph provides a natural way to model complex high-order relations, while its potentials for improving social recommendation are under-explored. In this paper, we fill this gap and propose a multi-channel hypergraph convolutional network to enhance social recommendation by leveraging high-order user relations. Technically, each channel in the network encodes a hypergraph that depicts a common high-order user relation pattern via hypergraph convolution. By aggregating the embeddings learned through multiple channels, we obtain comprehensive user representations to generate recommendation results. However, the aggregation operation might also obscure the inherent characteristics of different types of high-order connectivity information. To compensate for the aggregating loss, we innovatively integrate self-supervised learning into the training of the hypergraph convolutional network to regain the connectivity information with hierarchical mutual information maximization. The experimental results on multiple real-world datasets show that the proposed model outperforms the SOTA methods, and the ablation study verifies the effectiveness of the multi-channel setting and the selfsupervised task. The implementation of our model is available via https://github.com/Coder-Yu/RecQ.

show abstract

Section: Self-supervised Learningmentioning

confidence: 99%

Self-Supervised Multi-Channel Hypergraph Convolutional Network for Social Recommendation

Yin

et al. 2021

Proceedings of the Web Conference 2021

272

119

View full text Add to dashboard Cite

show abstract

“…There are several other proposed variants of GNNs; however, they are all confined to only capturing low-order graph structures around every node (Li et al, 2018). GNN models have recently incorporated graphlets (Tu et al, 2018;Feng & Chen, 2020), motifs (Zhao et al, 2018;Sankar et al, 2020;Subramonian, 2021), and anonymous walks (Long et al, 2020;Jin et al, 2020) to leverage higher-order graph structures. gl-DCNN (Tu et al, 2018) concatenates node graphlet information and node features for input to the diffusion-convolutional neural networks.…”

Section: Network Embeddingmentioning

confidence: 99%

SAlign: A Graph Neural Attention Framework for Aligning Structurally Heterogeneous Networks

Saxena

Chandra²

2023

jair

View full text Add to dashboard Cite

Network alignment techniques that map the same entities across multiple networks assume that the mapping nodes in two different networks have similar attributes and neighborhood proximity. However, real-world networks often violate such assumptions, having diverse attributes and structural properties. Node mapping across such structurally heterogeneous networks remains a challenge. Although capturing the nodes’ entire neighborhood (in low-dimensional embeddings) may help deal with these characteristic differences, the issue of over-smoothing in the representations that come from higherorder learning still remains a major problem. To address the above concerns, we propose SAlign: a supervised graph neural attention framework for aligning structurally heterogeneous networks that learns the correlation of structural properties of mapping nodes using a set of labeled (mapped) anchor nodes. SAlign incorporates nodes’ graphlet information with a novel structure-aware cross-network attention mechanism that transfers the required higher-order structure information across networks. The information exchanged across networks helps in enhancing the expressivity of the graph neural network, thereby handling any potential over-smoothing problem. Extensive experiments on three real datasets demonstrate that SAlign consistently outperforms the state-of-the-art network alignment methods by at least 1.3-8% in terms of accuracy score. The code is available at https://github.com/shruti400/SAlign for reproducibility.

show abstract

“…On the other hand, capturing high-order information based on graph neural networks have led to a renewed interest, of which research on motifs and hypergraphs is particularly important. (Sankar et al, 2020) learns statistical dependencies between structurally similar nodes with co-varying attributes and independent of network proximity, it maximizes motif-based mutual information, and dynamically prioritizes the significance of different motifs to learn network embedding. (Chen et al, 2021) proposed redundancy minimization among motifs which compares the motifs with each other and distills the features unique to each motif.…”

Section: Related Workmentioning

confidence: 99%

Capturing High-order Structures on Motif-based Graph Nerual Networks

Zhang¹

2022

Preprint

View full text Add to dashboard Cite

Graph Nerual Networks (GNNs) are effective models in graph embedding. It extracts shallow features and neighborhood information by aggregating neighbor information to learn the embedding representation of different nodes. However, the local topology information of many nodes in the network is similar, the network obtained by shallow embedding represents the network that is susceptible to structural noise, and the low-order embedding cannot capture the high-order network structure; on the other hand, the deep embedding undergoes multi-layer convolution. After the filters are stacked, the embedded distribution is destroyed, and graph smoothing occurs. To address these challenges, we propose a new framework that leverages network motifs to learn deep features of the network from low-level embeddings under the assumption of network homogeneity and transitivity, and then combines local neighborhood information with deeper global information fusion results in accurate representation of nodes. On real datasets, our model achieves significant performance improvement in link prediction and node classification.

show abstract

Beyond Localized Graph Neural Networks: An Attributed Motif Regularization Framework

Cited by 15 publications

References 32 publications

Self-Supervised Multi-Channel Hypergraph Convolutional Network for Social Recommendation

Self-Supervised Multi-Channel Hypergraph Convolutional Network for Social Recommendation

SAlign: A Graph Neural Attention Framework for Aligning Structurally Heterogeneous Networks

Capturing High-order Structures on Motif-based Graph Nerual Networks

Contact Info

Product

Resources

About