Graph Sampling-based Meta-Learning for Molecular Property Prediction

Authors: Xiang Zhuang, Qiang Zhang, Bin Wu, Keyan Ding, Yin Fang, Huajun Chen

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

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on 5 commonlyused benchmarks show GS-Meta consistently outperforms state-of-the-art methods by 5.71%-6.93% in ROC-AUC and verify the effectiveness of each proposed module.
Researcher Affiliation Collaboration 1College of Computer Science and Technology, Zhejiang University 2ZJU-Hangzhou Global Scientific and Technological Innovation Center 3Alibaba-Zhejiang University Joint Research Institute of Frontier Technologies
Pseudocode Yes Algorithm 1 Training and optimization algorithm.
Open Source Code Yes Our code is available at https: //github.com/HICAI-ZJU/GS-Meta.
Open Datasets Yes We use five common few-shot molecular property prediction datasets from the Molecule Net [Wu et al., 2018].
Dataset Splits No The paper defines the training set Dtrain and testing set Dtest for tasks, and how episodes are constructed with support and query sets (2-way K-shot), but it does not specify a separate, explicit validation dataset split for the overall benchmark datasets.
Hardware Specification No The paper acknowledges "Hangzhou AI Computing Center for their technical support" but does not provide specific details such as GPU models, CPU types, or other hardware specifications used for experiments.
Software Dependencies No The paper does not provide specific software dependencies with version numbers (e.g., Python, PyTorch, TensorFlow versions or specific library versions).
Experiment Setup Yes Appendix A.2 Training Details: We use Adam optimizer with learning rate 0.001. βinner = 0.01 and βouter = 0.001. The batch size is 32. The number of iterations for the inner loop is 1. The number of iterations for the outer loop is 1. We train for 100 epochs. τ = 0.5. For the graph neural network, we use a 2-layer GNN. The hidden dimension is 128.