arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1811.06231cond-mat.mtrl-scics.LG
收藏

聚合物性能预测的图卷积神经网络

Graph Convolutional Neural Networks for Polymers Property Prediction

Institute for Infocomm Research, A*STAR(新加坡科技研究局资讯通信研究院);Institute of Materials Research and Engineering(材料研究与工程研究院)

Minggang Zeng, Jatin Nitin Kumar, Zeng Zeng, Ramasamy Savitha, Vijay Ramaseshan Chandrasekhar, Kedar Hippalgaonkar

摘要:

一个快速准确的聚合物性能预测工具是苛刻的,将铺平道路,迭代反设计。在这项工作中,我们应用图卷积神经网络(GCNN)来预测聚合物的介电常数和能带隙。使用密度泛函理论(DFT)计算的性质作为地面真理,GCNN可以实现与DFT结果的显着一致。此外,我们还证明了GCNN优于其他机器学习算法。我们的工作证明,GCNN仅依赖于聚合物的形态数据,并消除了对复杂的手工描述符的要求,同时仍然提供快速预测的准确性。

英文摘要:

A fast and accurate predictive tool for polymer properties is demanding and will pave the way to iterative inverse design. In this work, we apply graph convolutional neural networks (GCNN) to predict the dielectric constant and energy bandgap of polymers. Using density functional theory (DFT) calculated properties as the ground truth, GCNN can achieve remarkable agreement with DFT results. Moreover, we show that GCNN outperforms other machine learning algorithms. Our work proves that GCNN relies only on morphological data of polymers and removes the requirement for complicated hand-crafted descriptors, while still offering accuracy in fast predictions.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑