Analysis of Atomistic Representations Using Weighted Skip-Connections
This work provides incremental insights into interpretability for machine learning models in computational chemistry.
The researchers extended the SchNet architecture with weighted skip connections to analyze the importance of different interaction blocks for molecular property prediction, finding that weightings depend strongly on chemical composition and molecular configuration.
In this work, we extend the SchNet architecture by using weighted skip connections to assemble the final representation. This enables us to study the relative importance of each interaction block for property prediction. We demonstrate on both the QM9 and MD17 dataset that their relative weighting depends strongly on the chemical composition and configurational degrees of freedom of the molecules which opens the path towards a more detailed understanding of machine learning models for molecules.