Communicative Representation Learning on Attributed Molecular Graphs

Authors: Ying Song, Shuangjia Zheng, Zhangming Niu, Zhang-hua Fu, Yutong Lu, Yuedong Yang

IJCAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments demonstrated that the proposed model obtained superior performances against state-of-the-art baselines on six chemical property datasets.
Researcher Affiliation Collaboration Ying Song1,2 , Shuangjia Zheng1 , Zhangming Niu2 , Zhang-Hua Fu3,4 , Yutong Lu1 and Yuedong Yang1 1Sun Yat-sen University 2Aladdin Healthcare Technologies Ltd 3The Chinese University of Hong Kong, Shenzhen 4Shenzhen Institute of Artificial Intelligence and Robotics for Society
Pseudocode Yes Algorithm 1 CMPNN embedding generation algorithm Input: Graph G(V, E); depth K; input node and edge features {xev,w, ev,w E, xv, v V }; aggregate function AGGREGATE; communicate function COMMUNICATE; weight matrix W. Output: Graph-wise vector representation z.
Open Source Code Yes https://github.com/SY575/CMPNN
Open Datasets Yes To enable head-to-head comparison of CMPNN to existing molecular representation methods, we evaluated our proposed model on six public benchmark datasets, including BBBP, Tox21, Clin Tox, and Sider for classification tasks, and ESOL and Freesolv for regression tasks.
Dataset Splits Yes Following [Yang et al.2019], we used a 5-fold cross validation and replicate experiments on each task for five times, and reported the mean and standard deviation of AUC or RMSE values. We evaluated all models on random and scaffold-based splits as recommended by [Wu et al.2018].
Hardware Specification Yes Our models were implemented by Pytorch and run on Ubuntu Linux 16 with NVIDIA Tesla V100 GPUs.
Software Dependencies No The paper mentions 'Pytorch', 'Ubuntu Linux 16', and 'RDKit' but does not specify version numbers for any of these software dependencies, which are necessary for full reproducibility.
Experiment Setup No The paper mentions applying "Bayesian Optimization to obtain the best hyperparameters of the models" but does not explicitly list the specific hyperparameter values (e.g., learning rate, batch size, number of epochs) or other detailed system-level training settings.