Integrating Explainability into Graph Neural Network Models for the Prediction of X-ray Absorption Spectra

Abstract

The use of sophisticated machine learning (ML) models, such as graph neural networks (GNNs), to predict complex molecular properties or all kinds of spectra has grown rapidly. However, ensuring the interpretability of these models' predictions remains a challenge. For example, a rigorous understanding of the predicted X-ray absorption spectrum (XAS) generated by such ML models requires an in-depth investigation of the respective black-box ML model used. Here, this is done for different GNNs based on a comprehensive, custom-generated XAS data set for small organic molecules. We show that a thorough analysis of the different ML models with respect to the local and global environments considered in each ML model is essential for the selection of an appropriate ML model that allows a robust XAS prediction. Moreover, we employ feature attribution to determine the respective contributions of various atoms in the molecules to the peaks observed in the XAS spectrum. By comparing this peak assignment to the core and virtual orbitals from the quantum chemical calculations underlying our data set, we demonstrate that it is possible to relate the atomic contributions via these orbitals to the XAS spectrum.

AI in computational chemistry through the lens of a decade-long journey

“…An inverse problem of elucidating structure from spectra is another important topic where AI is starting to become the center of research. 161,162…”

Gaining More Insights from Synchrotron-Based X-ray Spectroscopy for Alkali Ion Rechargeable Batteries

“…Benefiting from advances in data science, the machine learning (ML) approach has emerged as an unprecedented tool for automatically handling large data sets. − For example, the combined use of attribute extraction and clustering algorithms allows the identification of unanticipated minority phases from over 10 million Co K-edge XANES spectra covering more than 100 LiCoO 2 particles collected in the STXM experiment (Figure b). Beyond the battery community, the field of condensed matter and catalysis has witnessed the significant power of ML in X-ray spectroscopy. ,,, For instance, the direct conversion of XANES data into the radial distribution function (RDF) was accomplished using an artificial neural network (ANN) ML model (Figure c). More recently, the incorporation of explainability into the graph neural network (GNN) architecture not only facilitates accurate predictions of XANES spectra but also enhances our current comprehension of XANES (Figure d).…”

Machine-learning strategies for the accurate and efficient analysis of x-ray spectroscopy

“…Using this, the authors demonstrated that the resulting network could predict spectra with 90% accuracy of the predicted spectral peak locations being within 1 eV of the expected energy, very comparable to the performance achieved by Rankine and Penfold [92], although this did not specifically take advantage of the message passing framework to encode higher-order information. A similar approach was recently adopted by Kotobi et al [126] in which the authors focused on developing an explainable network. Indeed, using feature attribution the authors were able to quantify the contribution of each atom to peaks in the spectrum, which subsequently could be compared to orbitals involved in the transitions.…”