Representing Polymers as Periodic Graphs with Learned Descriptors for Accurate Polymer Property Predictions
This work addresses the problem of polymer property prediction for materials science, offering a novel representation that improves accuracy over hand-designed methods, though it is incremental in advancing graph-based approaches for polymers.
The paper tackled the challenge of accurately representing complex polymer structures for machine learning by developing a periodic polymer graph representation that accounts for periodicity, combined with a message-passing neural network to learn descriptors automatically, resulting in a 20% average reduction in prediction error across 10 polymer properties.
One of the grand challenges of utilizing machine learning for the discovery of innovative new polymers lies in the difficulty of accurately representing the complex structures of polymeric materials. Although a wide array of hand-designed polymer representations have been explored, there has yet to be an ideal solution for how to capture the periodicity of polymer structures, and how to develop polymer descriptors without the need for human feature design. In this work, we tackle these problems through the development of our periodic polymer graph representation. Our pipeline for polymer property predictions is comprised of our polymer graph representation that naturally accounts for the periodicity of polymers, followed by a message-passing neural network (MPNN) that leverages the power of graph deep learning to automatically learn chemically-relevant polymer descriptors. Across a diverse dataset of 10 polymer properties, we find that this polymer graph representation consistently outperforms hand-designed representations with a 20% average reduction in prediction error. Our results illustrate how the incorporation of chemical intuition through directly encoding periodicity into our polymer graph representation leads to a considerable improvement in the accuracy and reliability of polymer property predictions. We also demonstrate how combining polymer graph representations with message-passing neural network architectures can automatically extract meaningful polymer features that are consistent with human intuition, while outperforming human-derived features. This work highlights the advancement in predictive capability that is possible if using chemical descriptors that are specifically optimized for capturing the unique chemical structure of polymers.