Unleashing the Power of Graph Data Augmentation on Covariate Distribution Shift

Authors: Yongduo Sui, Qitian Wu, Jiancan Wu, Qing Cui, Longfei Li, Jun Zhou, Xiang Wang, Xiangnan He

NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments with in-depth empirical analysis demonstrate the superiority of our approach. The implementation codes are publicly available at https://github.com/yongduosui/AIA. In this section, we conduct extensive experiments to answer the following Research Questions: ... We first make comparisons with various baselines in Table 1, and have the following observations: ... We also conduct extensive experiments and in-depth analyses.
Researcher Affiliation Collaboration 1University of Science and Technology of China, 2Shanghai Jiao Tong University, 3Ant Group
Pseudocode Yes Algorithm 1: Estimation of Graph Covariate shift; Algorithm 2: Adversarial Invariant Augmentation
Open Source Code Yes The implementation codes are publicly available at https://github.com/yongduosui/AIA.
Open Datasets Yes We use graph OOD datasets [2] and OGB datasets [20], which include Motif, CMNIST, Molbbbp, and Molhiv.
Dataset Splits Yes For size covariate shift, we use small-size of graphs for training, while the validation and the test sets include the middle- and the large-size graphs, respectively. ... For scaffold shift, we follow [2] and use scaffold split to create training, validation and test sets.
Hardware Specification Yes We use the NVIDIA Ge Force RTX 3090 (24GB GPU) to conduct all our experiments.
Software Dependencies No The paper mentions using GIN as the backbone and discusses some general software aspects but does not provide specific version numbers for software dependencies (e.g., Python, PyTorch, or other libraries).
Experiment Setup Yes We tune the hyper-parameters in the following ranges: α and β {0.01,0.005,0.001}; λ2 {0.1,...,0.9}; γ {0.01,0.1,0.2,0.5,1.0,1.5,2.0,3.0,5.0}. The hyper-parameters are summarized in Table 5.