MUBen: Benchmarking the Uncertainty of Molecular Representation Models

arXiv:2306.10060v3 Announce Type: replace-cross
Abstract: Large molecular representation models pre-trained on massive unlabeled data have shown great success in predicting molecular properties. However, these models may tend to overfit the fine-tuning data, resulting in over-confident predictions on test data that fall outside of the training distribution. To address this issue, uncertainty quantification (UQ) methods can be used to improve the models' calibration of predictions. Although many UQ approaches exist, not all of them lead to improved performance. While some studies …

abstract arxiv benchmarking cs.lg data distribution fine-tuning however issue massive physics.chem-ph predictions quantification representation success test training type uncertainty