HAGO-Net: Hierarchical Geometric Message Passing for Molecular Representation Learning
Authors: Hongbin Pei, Taile Chen, Chen A, Huiqi Deng, Jing Tao, Pinghui Wang, Xiaohong Guan
AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The proposed models are validated by extensive comparisons on four challenging benchmarks. Notably, the models exhibited state-of-the-art performance in molecular chirality identification and property prediction, achieving state-of-the-art performance on five properties of QM9 dataset. The models also achieved competitive results on molecular dynamics prediction task. |
| Researcher Affiliation | Academia | Hongbin Pei1, Taile Chen1, Chen A1, Huiqi Deng2, Jing Tao1*, Pinghui Wang1, Xiaohong Guan1 1MOE KLINNS Lab, Xi an Jiaotong University, China 2Shanghai Jiao Tong University, China peihongbin@xjtu.edu.cn, stardust@stu.xjtu.edu.cn |
| Pseudocode | No | The paper describes the proposed HAGO-MPS with definitions and mathematical formulations but does not include any explicitly labeled pseudocode or algorithm blocks. |
| Open Source Code | Yes | Model configurations in all experiments are detailed in our Git Hub repository (Git-repo 2023). |
| Open Datasets | Yes | We empirically validate and analyze HAGO-Net and DHAGO-Net on four challenging molecular datasets, including MD17, QM9, GEOM-QM9, and Pub Chem3D. |
| Dataset Splits | Yes | We adopt the same dataset split in (Liu et al. 2022), i.e., 110,000 molecules for training, 10,000 molecules for validation, and the rest for testing. |
| Hardware Specification | Yes | All experiments are conducted on eight NVIDIA Tesla V100 GPUs. |
| Software Dependencies | No | The paper does not explicitly list specific software dependencies with version numbers (e.g., Python, PyTorch, TensorFlow versions). |
| Experiment Setup | No | Model configurations in all experiments are detailed in our Git Hub repository (Git-repo 2023), but specific hyperparameters or detailed training settings are not provided within the main text of the paper. |