AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Injecting Domain Knowledge from Empirical Interatomic Potentials to Neural Networks for Predicting Material Properties

Zeren Shui, Daniel S. Karls, Mingjian Wen, ilia Andreyevich Nikiforov, Ellad Tadmor, George Karypis

NeurIPS 2022 · Label-efficient and robust learning › Recognition and long-tail · approach: Dense features and foundation priors

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add label augmentation with EIP labels (ablation)SchNet · ANI-Alconfiguration-level MAE (eV): 0.1693 -> 0.0845Improved
add label augmentation (ablation)SOAPNet · KIM-Siconfiguration-level MAE (eV): 0.7706 -> 0.5595Improved
add label augmentation (ablation)CGCNN · AgAuconfiguration-level MAE (eV): 1.6683 -> 1.6065Improved
add multi-task pretraining on EIP energies (ablation)SchNet · AgAuconfiguration-level MAE (eV): 0.7290 -> 0.3353Improved
add multi-task pretraining (ablation)GemNet · KIM-Siconfiguration-level MAE (eV): 0.5138 -> 0.5024Improved
combine multi-task pretraining and label augmentation (ablation)SchNet · KIM-Siconfiguration-level MAE (eV): 0.4805 -> 0.3719Improved
combine both strategies vs multi-task pretraining alone (ablation)GemNet · AgAuconfiguration-level MAE (eV): 0.5057 -> 0.6074Hurt
combine both strategies vs multi-task pretraining alone (ablation)SchNet · AgAuconfiguration-level MAE (eV): 0.3353 -> 0.3496Hurt
replace Tukey loss with MSE loss (ablation)SchNet +LA · ANI-Alconfiguration-level MAE (eV): 0.0845 -> 0.1031Hurt
replace Tukey loss with MSE loss (ablation)SOAPNet +LA · KIM-Siconfiguration-level MAE (eV): 0.5595 -> 0.5556No clear effect
label augmentation with only the worst EIP (ablation)SOAPNet · KIM-Siconfiguration-level MAE (eV): 0.7582 -> 0.766Hurt
label augmentation with only the best EIP vs default 8 EIPs (ablation)SOAPNet · KIM-Siconfiguration-level MAE (eV): 0.5673 -> 0.6084Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy