GPS++: Reviving the Art of Message Passing for Molecular Property Prediction
This work addresses molecular property prediction for computational chemistry, offering an incremental improvement by integrating existing methods to enhance accuracy, especially when 3D positional information is unavailable.
The authors tackled molecular property prediction by developing GPS++, a hybrid Message Passing Neural Network/Graph Transformer model, achieving state-of-the-art results on the PCQM4Mv2 dataset and showing that message passing remains competitive without global self-attention.
We present GPS++, a hybrid Message Passing Neural Network / Graph Transformer model for molecular property prediction. Our model integrates a well-tuned local message passing component and biased global attention with other key ideas from prior literature to achieve state-of-the-art results on large-scale molecular dataset PCQM4Mv2. Through a thorough ablation study we highlight the impact of individual components and find that nearly all of the model's performance can be maintained without any use of global self-attention, showing that message passing is still a competitive approach for 3D molecular property prediction despite the recent dominance of graph transformers. We also find that our approach is significantly more accurate than prior art when 3D positional information is not available.