Uncertainty Quantification in Graph Neural Networks with Shallow Ensembles
This work addresses the robustness issue in GNN-based materials modeling for researchers and practitioners, though it appears incremental as it applies an existing UQ method to a specific domain.
The paper tackled the problem of unreliable predictions in Graph Neural Networks (GNNs) for materials discovery when encountering out-of-domain data by exploring Uncertainty Quantification (UQ) techniques, specifically Direct Propagation of Shallow Ensembles (DPOSE), and found that DPOSE successfully distinguishes between in-domain and out-of-domain samples with higher uncertainty for unobserved classes.
Machine-learned potentials (MLPs) have revolutionized materials discovery by providing accurate and efficient predictions of molecular and material properties. Graph Neural Networks (GNNs) have emerged as a state-of-the-art approach due to their ability to capture complex atomic interactions. However, GNNs often produce unreliable predictions when encountering out-of-domain data and it is difficult to identify when that happens. To address this challenge, we explore Uncertainty Quantification (UQ) techniques, focusing on Direct Propagation of Shallow Ensembles (DPOSE) as a computationally efficient alternative to deep ensembles. By integrating DPOSE into the SchNet model, we assess its ability to provide reliable uncertainty estimates across diverse Density Functional Theory datasets, including QM9, OC20, and Gold Molecular Dynamics. Our findings often demonstrate that DPOSE successfully distinguishes between in-domain and out-of-domain samples, exhibiting higher uncertainty for unobserved molecule and material classes. This work highlights the potential of lightweight UQ methods in improving the robustness of GNN-based materials modeling and lays the foundation for future integration with active learning strategies.