Feature-dependent graph convolutional autoencoders with adversarial training methods
Paper
Paper/Presentation Title | Feature-dependent graph convolutional autoencoders with adversarial training methods |
---|---|
Presentation Type | Paper |
Authors | Wu, Di, Hu, Ruiqi, Zheng, Yu, Jiang, Jing, Sharma, Nabin and Blumenstein, Michael |
Journal or Proceedings Title | Proceedings of 2019 International Joint Conference on Neural Networks (IJCNN) |
Number of Pages | 8 |
Year | 2019 |
Publisher | IEEE (Institute of Electrical and Electronics Engineers) |
Place of Publication | United States |
Digital Object Identifier (DOI) | https://doi.org/10.1109/IJCNN.2019.8852314 |
Web Address (URL) of Paper | https://ieeexplore.ieee.org/abstract/document/8852314 |
Web Address (URL) of Conference Proceedings | https://ieeexplore.ieee.org/xpl/conhome/8840768/proceeding |
Conference/Event | 2019 International Joint Conference on Neural Networks (IJCNN) |
Event Details | 2019 International Joint Conference on Neural Networks (IJCNN) Parent International Joint Conference on Neural Networks (IJCNN) Delivery In person Event Date 14 to end of 19 Jul 2019 Event Location Budapest, Hungary |
Abstract | Graphs are ubiquitous for describing and modeling complicated data structures, and graph embedding is an effective solution to learn a mapping from a graph to a low-dimensional vector space while preserving relevant graph characteristics. Most existing graph embedding approaches either embed the topological information and node features separately or learn one regularized embedding with both sources of information, however, they mostly overlook the interdependency between structural characteristics and node features when processing the graph data into the models. Moreover, existing methods only reconstruct the structural characteristics, which are unable to fully leverage the interaction between the topology and the features associated with its nodes during the encoding-decoding procedure. To address the problem, we propose a framework using autoencoder for graph embedding (GED) and its variational version (VEGD). The contribution of our work is two-fold: 1) the proposed frameworks exploit a feature-dependent graph matrix (FGM) to naturally merge the structural characteristics and node features according to their interdependency; and 2) the Graph Convolutional Network (GCN) decoder of the proposed framework reconstructs both structural characteristics and node features, which naturally possesses the interaction between these two sources of information while learning the embedding. We conducted the experiments on three real-world graph datasets such as Cora, Citeseer and PubMed to evaluate our framework and algorithms, and the results outperform baseline methods on both link prediction and graph clustering tasks. |
Keywords | Graph Embedding; Graph Convolutional Neural Networks; Generative Adversarial Network |
Contains Sensitive Content | Does not contain sensitive content |
ANZSRC Field of Research 2020 | 4602. Artificial intelligence |
460502. Data mining and knowledge discovery | |
Public Notes | Files associated with this item cannot be displayed due to copyright restrictions, but may be accessed online. Please see the link in the URL field. |
Byline Affiliations | University of Technology Sydney |
Northwest A&F University, China |
https://research.usq.edu.au/item/z4y20/feature-dependent-graph-convolutional-autoencoders-with-adversarial-training-methods
46
total views0
total downloads3
views this month0
downloads this month