Adversarial Network Embedding

Authors: Quanyu Dai, Qiang Li, Jian Tang, Dan Wang

AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental As shown by the empirical results, our method is competitive with or superior to state-of-the-art approaches on benchmark network embedding tasks.
Researcher Affiliation Academia 1Department of Computing, The Hong Kong Polytechnic University, Hong Kong 2School of Software, FEIT, The University of Technology Sydney, Australia 3HEC Montreal, Canada 4Montreal Institute for Learning Algorithms, Canada
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks.
Open Source Code No The source code will be available online.
Open Datasets Yes Cora and Citeseer are paper citation networks constructed by (Mc Callum et al. 2000). Wiki (Sen et al. 2008) is a network... Cit-DBLP is a paper citation network extracted from DBLP dataset (Tang et al. 2008).
Dataset Splits No The paper states 'We range the training ratio from 10% to 90% for comprehensive evaluation' but does not specify validation splits or ratios explicitly.
Hardware Specification No The paper does not provide specific details regarding the hardware used for running experiments.
Software Dependencies No All experiments are carried out with support vector classifier in Liblinear package (Fan et al. 2008). The paper mentions Liblinear but does not provide its version number.
Experiment Setup Yes For both Deep Walk and node2vec, the window size s, the walk length l and the number of walks η per node are set to 10, 80 and 10, respectively, for fair comparison. ... Specifically, the generator is a single-layer network with leaky Re LU activations (with a leak of 0.2) and batch normalization ... The number of negative samples K is set to 5... We use RMSProp optimizer with learning rate as 0.001.