Revisiting Self-Supervised Heterogeneous Graph Learning from Spectral Clustering Perspective

Authors: YUJIE MO, Zhihe Lu, Runpeng Yu, Xiaofeng Zhu, Xinchao Wang

NeurIPS 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results affirm the superiority of our method, showcasing remarkable improvements in several downstream tasks compared to existing methods.
Researcher Affiliation Academia 1School of Computer Science and Engineering, University of Electronic Science and Technology of China 2National University of Singapore
Pseudocode Yes Algorithm 1 The pseudo-code of the proposed method.
Open Source Code Yes The code of the proposed method is released at https://github.com/Yujie Mo/SCHOOL.
Open Datasets Yes We use four public heterogeneous graph datasets and two public homogeneous graph datasets from various domains. Heterogeneous graph datasets include three academic datasets (i.e., ACM [56], DBLP [56], and Aminer [11]), and one business dataset (i.e., Yelp [27]). Homogeneous graph datasets include two sale datasets (i.e., Photo and Computers [43]).
Dataset Splits No Table 3 provides '#Training' and '#Test' splits for the datasets but does not explicitly mention 'validation' splits.
Hardware Specification Yes All experiments were implemented in Py Torch and conducted on a server with 8 NVIDIA Ge Force 3090 (24GB memory each).
Software Dependencies No The paper mentions that experiments were 'implemented in Py Torch' but does not specify a version number for Py Torch or any other software dependencies.
Experiment Setup Yes In the proposed method, all parameters were optimized by the Adam optimizer [19] with an initial learning rate. Moreover, We use early stopping with a patience of 30 to train the proposed SHGL model. We report the settings for the dimensions of encoders in Table 5.