Benchmarking Graphormer on Large-Scale Molecular Modeling Datasets
This work provides incremental improvements for molecular modeling researchers, enhancing performance on quantum chemistry and catalyst-adsorbate reaction tasks.
The authors improved Graphormer with architecture modifications and adaptation to 3D molecular dynamics, achieving better performance on large-scale molecular modeling datasets, including much lower MAE on the PCQM4M dataset and outperforming competitors in the Open Catalyst Challenge.
This technical note describes the recent updates of Graphormer, including architecture design modifications, and the adaption to 3D molecular dynamics simulation. With these simple modifications, Graphormer could attain better results on large-scale molecular modeling datasets than the vanilla one, and the performance gain could be consistently obtained on 2D and 3D molecular graph modeling tasks. In addition, we show that with a global receptive field and an adaptive aggregation strategy, Graphormer is more powerful than classic message-passing-based GNNs. Empirically, Graphormer could achieve much less MAE than the originally reported results on the PCQM4M quantum chemistry dataset used in KDD Cup 2021. In the meanwhile, it greatly outperforms the competitors in the recent Open Catalyst Challenge, which is a competition track on NeurIPS 2021 workshop, and aims to model the catalyst-adsorbate reaction system with advanced AI models. All codes could be found at https://github.com/Microsoft/Graphormer.