Triplet Enhanced AutoEncoder: Model-free Discriminative Network Embedding

Authors: Yao Yang, Haoran Chen, Junming Shao

IJCAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive empirical results on three public datasets (i.e, Cora, Citeseer and Blog Catalog) show that TEA is stable and achieves state-of-the-art performance compared with both supervised and unsupervised network embedding approaches on various percentages of labeled data.
Researcher Affiliation Academia Yao Yang , Haoran Chen and Junming Shao Data Mining Lab, University of Electronic Science and Technology of China {yaoyang, haorchen}@std.uestc.edu.cn, junmshao@uestc.edu.cn
Pseudocode No The paper describes the model architecture and optimization process but does not include a structured pseudocode or algorithm block.
Open Source Code Yes The source code can be obtained from https://github.com/yybeta/TEA.
Open Datasets Yes Cora1: This is a citation network of academic papers... 1http://www.cs.umd.edu/ sen/lbc-proj/data/cora.tgz; Citeseer2: This is another research paper set... 2http://www.cs.umd.edu/ sen/lbc-proj/data/citeseer.tgz; Blog Catalog3: This is a network of social relationships between users on the Blog Catalog website... 3http://socialcomputing.asu.edu/datasets/Blog Catalog3
Dataset Splits Yes For the node classification task, we adopted micro-F1 and macro-F1... Here four-fold cross-validation is applied... We increase the labeled ratio from 20% to 80%, to compare the performance of different algorithms.
Hardware Specification No The paper mentions training models but does not provide specific hardware details (e.g., CPU/GPU models, memory) used for the experiments.
Software Dependencies No The paper mentions using Adam optimizer and Lib Linear, but does not provide specific version numbers for these or other software dependencies.
Experiment Setup Yes The dimension of each layer is listed in Table 1. The hyper-parameters of α and β are tuned by using grid search on the validation set, the sensitivity is further evaluated in Section 4.5. The models were trained for 20 epochs with Adam optimizer. We set the representation dimension to 100 for baseline algorithms, same as our model. The parameters in baselines are tuned. For Gra Rep, we set the maximum matrix transition step K = 5 for Blog Catalog, K = 3 for Cora and Citeseer. For node2vec, we set the window size as 10, the walk length as 80, the walk per vertex as 20, both in-out and return hyper-parameters are set to 0.25. For DNGR, we set the neural network structures the same as TEA. For MMDW, we set the gradient balance parameter η = 10 2.