Prediction rigidities for atomistic ML models

Add to your list(s) Download to your calendar using vCal

• Sanggyu Chong, EPFL
• Monday 09 September 2024, 14:30-15:30
• Yusuf Hamied Department of Chemistry, Wolfson Lecture Theatre.

If you have a question about this talk, please contact Alexander R Epstein.

The widespread application of atomistic machine learning (ML) in the chemical sciences has made it very important to understand how the models learn to correlate chemical structures with their properties, and what can be done to improve the training efficiency whilst guaranteeing interpretability and transferability. In this work, we demonstrate the wide utility of prediction rigidities, a family of metrics derived from the loss function, in understanding the robustness of atomistic ML model predictions. We show that the prediction rigidities allow the assessment of the model not only at the global level for uncertainty quantification, but also on the local or the component-wise level at which the intermediate (e.g. atomic, body-ordered, or range-separated) predictions are made. This allows for an understanding of the model’s learning behavior, and helps to guide efficient dataset construction for training.

This talk is part of the Lennard-Jones Centre series.

This talk is included in these lists:

• Hanchen DaDaDash
• Lennard-Jones Centre
• Yusuf Hamied Department of Chemistry, Wolfson Lecture Theatre

Note that ex-directory lists are not shown.