Unsupervised Attributed Multiplex Network Embedding

By Chanyoung Park, Donghyun Kim, Jiawei Han, Hwanjo Yu. AAAI Conference on Artificial Intelligence, 2019

341 citations recorded by the source corpus.

Nodes in a multiplex network are connected by multiple types of relations. However, most existing network embedding methods assume that only a single type of relation exists between nodes. Even for those that consider the multiplexity of a network, they overlook node attributes, resort to node labels for training, and fail to model the global properties of a graph. We present a simple yet effective unsupervised network embedding method for attributed multiplex network called DMGI, inspired by Deep Graph Infomax (DGI) that maximizes the mutual information between local patches of a graph, and the global representation of the entire graph. We devise a systematic way to jointly integrate the node embeddings from multiple graphs by introducing 1) the consensus regularization framework that minimizes the disagreements among the relation-type specific node embeddings, and 2) the universal discriminator that discriminates true samples regardless of the relation types. We also show that the attention mechanism infers the importance of each relation type, and thus can be useful for filtering unnecessary relation types as a preprocessing step. Extensive experiments on various downstream tasks demonstrate that DMGI outperforms the state-of-the-art methods, even though DMGI is fully unsupervised.

Subfields: Artificial Intelligence, Statistical and Nonlinear Physics

View Paper: https://www.semanticscholar.org/paper/ebf27510ccbe7a5e5edab92ddc135d2552c8df00

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Philip S. Yu
2. Chuan Shi
3. Xiao Wang
4. Shirui Pan
5. Jiawei Han
6. Wenwu Zhu
7. Chuan Zhou
8. Heng Ji

In Statistical and Nonlinear Physics

1. Chuan Shi
2. Philip S. Yu
3. Xiao Wang
4. Shirui Pan
5. Jiawei Han
6. Ting-Li Chen
7. Wenwu Zhu
8. Zhangyang Wang

Citations in the corpus

• Self-Supervised Learning on Graphs: Contrastive, Generative, or Predictive (2021, background): “DMGI [80] for multiplexed graphs, and BiGI [37] for bipartite graphs.”
• Self-supervised Heterogeneous Graph Neural Network with Co-contrastive Learning (2021, methodology): “In heterogeneous domain, DMGI [24] conducts contrastive learning between original network and corrupted network on each single view, meta-path, and designs a consensus regularization to guide the fusion of different meta-paths.”
• Graph Contrastive Learning with Augmentations (2020, methodology): “choose f 1 = f 2 and generate G^ i;G^ jthrough data augmentation, while with various choices of the compositions result in (4) instantiating as other specific contrastive learning algorithms including [54, 55, 56, 21, 57, 58, 59] also shown in in Appendix F. 4 The Role of Data Augmentation in Graph C”
• A Comprehensive Survey on Community Detection With Deep Learning (2021, methodology): “As an example, deep graph infomax for attributed multiplex network embedding (DMGI) [66] independently embeds each relationship type and computes the network embeddings to maximize globally shared features when detecting communities.”
• Large Language Models on Graphs: A Comprehensive Survey (2023, background): “METERN [75] adds learnable relation embeddings to node textual sequences for text-based multiplex representation learning on graphs [92].”
• Bootstrapped Representation Learning on Graphs (2021, neutral): “DGI was also generalized to spatiotemporal graphs by ST-DGI (Opolka et al., 2019), and multiplex networks by DMGI (Park et al., 2020).”
• Graph Self-Supervised Learning: A Survey (2021, background): “Apart from those methods relying on carefully-crafted negative samples, approaches like BGRL [84] propose to contrast on graph instances themselves and thus alleviate TABLE 3 Main Characteristics of Contrast-Based Graph SSL Approaches Approach Pretext Task Category Downstream Task Level Training Sch”
• Graph Contrastive Learning Automated (2021, background): “…representations from unlabeled graph data (Veli ˇ ckovi ´ c et al., 2018; Sun et al., 2019; You et al., 2020a; Qiu et al., 2020; Hassani & Khasahmadi, 2020; Zhu et al., 2020b;c; Chen et al., 2020b;a; Ren et al., 2019; Park et al., 2020; Peng et al., 2020b; Jin et al., 2021a; Wang & Liu, 2021).”
• Implicit Graph Neural Networks (2020, methodology): “Given the same experimental settings, we reuse the results of baselines from (Park et al., 2019).”