Beyond Localized Graph Neural Networks: An Attributed Motif Regularization Framework

  title={Beyond Localized Graph Neural Networks: An Attributed Motif Regularization Framework},
  author={Aravind Sankar and Junting Wang and Adit Krishnan and H. Sundaram},
  journal={2020 IEEE International Conference on Data Mining (ICDM)},
We present InfoMotif, a new semi-supervised, motif-regularized, learning framework over graphs. We overcome two key limitations of message passing in popular graph neural networks (GNNs): localization (a $k$-layer GNN cannot utilize features outside the $k$-hop neighborhood of the labeled training nodes) and over-smoothed (structurally indistinguishable) representations. We propose the concept of attributed structural roles of nodes based on their occurrence in different network motifs… 

Figures and Tables from this paper

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

This paper proposes a multi-channel hypergraph convolutional network to enhance social recommendation by leveraging high-order user relations and innovatively integrates self-supervised learning into the training of the hypergraph Convolutional Network to regain the connectivity information with hierarchical mutual information maximization.

Graph Learning with 1D Convolutions on Random Walks

It is demonstrated empirically that CRAWL matches or outperforms state-of-the-art GNN architectures across a multitude of benchmark datasets for classification and regression on graphs.

Motif Graph Neural Network

Motif Graph Neural Network is proposed, a novel framework to better capture high-order structures, hinging on the proposed motif redundancy minimization operator and injective motif combination, and it is demonstrated that MGNN outperforms state-of-the-art methods on seven public benchmarks on both node classification and graph classification tasks.

A Short Survey on Inductive Biased Graph Neural Networks

A new taxonomy is proposed which divides general GNNs into recurrent GNNS, spectral GNN's, spatial GNN’s and topology-aware GNNes, with a focus on the latest developments in graph representation learning.

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

A new framework is proposed 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.

Self-Supervised Graph Neural Networks via Diverse and Interactive Message Passing

A novel Diverse and Interactive Message Passing (DIMP) is proposed for self-supervised learning by overcoming over-smoothing issue and is equivalent to performing an overlapping community detection via expectation-maximization (EM).

Bilateral Filtering Graph Convolutional Network for Multi-relational Social Recommendation in the Power-law Networks

A novel Bilateral Filtering Heterogeneous Attention Network (BFHAN) is proposed, which improves long-tail node representations and leverages multi-typed social relationships between user nodes and outperforms several state-of-the-art GCN-based methods for social recommendation tasks.

Graph Neural Networks for Friend Ranking in Large-scale Social Platforms

This work designs a neural architecture GraFRank to learn expressive user representations from multiple feature modalities and user-user interactions, and employs modality-specific neighbor aggregators and cross-modality attentions to learn multi-faceted user representations.

Ranking User-Generated Content via Multi-Relational Graph Convolution

A modular platform-agnostic framework to represent the contrastive and similarity-based relational aspects of user-generated content via independently induced content graphs and a surprising result-applying classical boosting techniques to combine final-layer embeddings across the content graphs significantly outperforms the typical stacking, fusion, or neighborhood embedding aggregation methods in graph convolutional architectures.

On the Equivalence of Decoupled Graph Convolution Network and Label Propagation

A new label propagation method named Propagation then Training Adaptively (PTA) is proposed, which overcomes the flaws of the decoupled GCN with a dynamic and adaptive weighting strategy.



A Structural Graph Representation Learning Framework

This work formulate higher-order network representation learning and describes a general framework called HONE for learning such structural node embeddings from networks via the subgraph patterns (network motifs, graphlet orbits/positions) in a nodes neighborhood.

node2vec: Scalable Feature Learning for Networks

In node2vec, an algorithmic framework for learning continuous feature representations for nodes in networks, a flexible notion of a node's network neighborhood is defined and a biased random walk procedure is designed, which efficiently explores diverse neighborhoods.

Representation Learning on Graphs with Jumping Knowledge Networks

This work explores an architecture -- jumping knowledge (JK) networks -- that flexibly leverages, for each node, different neighborhood ranges to enable better structure-aware representation in graphs.

Graph Convolutional Networks with Motif-based Attention

This work proposes a motif-based graph attention model, called Motif Convolutional Networks, which generalizes past approaches by using weighted multi-hop motif adjacency matrices to capture higher-order neighborhoods.

Position-aware Graph Neural Networks

Position-aware Graph Neural Networks (P-GNNs) are proposed, a new class of GNNs for computing position-aware node embeddings that are inductive, scalable, and can incorporate node feature information.

DEMO-Net: Degree-specific Graph Neural Networks for Node and Graph Classification

A generic degree-specific graph neural network named DEMO-Net motivated by Weisfeiler-Lehman graph isomorphism test that recursively identifies 1-hop neighborhood structures is proposed and two multi-task learning methods: degree- specific weight and hashing functions for graph convolution are designed.

Learning Structural Node Embeddings via Diffusion Wavelets

GraphWave is developed, a method that represents each node's network neighborhood via a low-dimensional embedding by leveraging heat wavelet diffusion patterns and mathematically proves that nodes with similar network neighborhoods will have similar GraphWave embeddings even though these nodes may reside in very different parts of the network, and the method scales linearly with the number of edges.

Graph Attention Networks

We present graph attention networks (GATs), novel neural network architectures that operate on graph-structured data, leveraging masked self-attentional layers to address the shortcomings of prior

Meta-GNN: Metagraph Neural Network for Semi-supervised learning in Attributed Heterogeneous Information Networks

This paper proposes a novel metagraph convolution operation to extract features from local metagraph-structured neighborhoods, thus capturing semantic higher-order relationships in AHINs and proposes a proposed neural architecture Meta-GNN, which extracts features of diverse semantics by utilizing multiple metagraphs.

motif2vec: Motif Aware Node Representation Learning for Heterogeneous Networks

A novel efficient algorithm, motif2vec that learns node representations or embeddings for heterogeneous networks, and is evaluated on multiple real-world networks from diverse domains and against existing state-of-the-art methods on multi-class node classification and link prediction tasks, and demonstrates its consistent superiority over prior work.