Graph Neural Network with Local Frame for Molecular Potential Energy Surface
This work addresses the challenge of efficiently capturing geometric information in molecular modeling for scientific applications, offering a simpler and more scalable approach.
The authors tackled the problem of modeling molecular potential energy surfaces by introducing a local frame method that converts equivariant features to invariant ones, enabling the use of ordinary Graph Neural Networks (GNNs) without complex symmetry designs. Their model achieved state-of-the-art accuracy with 30% inference time and 10% GPU memory usage compared to baselines.
Modeling molecular potential energy surface is of pivotal importance in science. Graph Neural Networks have shown great success in this field. However, their message passing schemes need special designs to capture geometric information and fulfill symmetry requirement like rotation equivariance, leading to complicated architectures. To avoid these designs, we introduce a novel local frame method to molecule representation learning and analyze its expressivity. Projected onto a frame, equivariant features like 3D coordinates are converted to invariant features, so that we can capture geometric information with these projections and decouple the symmetry requirement from GNN design. Theoretically, we prove that given non-degenerate frames, even ordinary GNNs can encode molecules injectively and reach maximum expressivity with coordinate projection and frame-frame projection. In experiments, our model uses a simple ordinary GNN architecture yet achieves state-of-the-art accuracy. The simpler architecture also leads to higher scalability. Our model only takes about 30% inference time and 10% GPU memory compared to the most efficient baselines.