文献类型：期刊文献

英文题名：Generalizing property prediction of ionic liquids from limited labeled data: a one-stop framework empowered by transfer learning

作者：Chen, Guzhong[1,2];Song, Zhen[1];Qi, Zhiwen[1];Sundmacher, Kai[2,3]

机构：[1]East China Univ Sci & Technol, Sch Chem Engn, State Key Lab Chem Engn, 130 Meilong Rd, Shanghai 200237, Peoples R China;[2]Max Planck Inst Dynam Complex Tech Syst, Proc Syst Engn, Sandtorstr 1, D-39106 Magdeburg, Germany;[3]Otto von Guericke Univ, Proc Syst Engn, Univ Pl 2, D-39106 Magdeburg, Germany

年份：2023

卷号：2

期号：3

起止页码：591

外文期刊名：DIGITAL DISCOVERY

收录：WOS:【ESCI(收录号:WOS:001101910500001)】；

基金：This research is supported by the National Natural Science Foundation of China (NSFC) under the grants of 22278134, 22208098, and 21CAA01709. G. C. acknowledges the financial support of the China Scholarship Council (CSC) for his joint PhD program (No. 202106740022) with the Max Planck Institute for Dynamics of Complex Technical Systems, Magdeburg, Germany.

语种：英文

摘要：Ionic liquids (ILs) could find use in almost every chemical process due to their wide spectrum of unique properties. The crux of the matter lies in whether a task-specific IL selection from enormous chemical space can be achieved by property prediction, for which limited labeled data represents a major obstacle. Here, we propose a one-stop ILTransR (IL transfer learning of representations) that employs large-scale unlabeled data for generalizing IL property prediction from limited labeled data. By first pre-training on similar to 10 million IL-like molecules, IL representations are derived from the encoder state of a transformer model. Employing the pre-trained IL representations, convolutional neural network (CNN) models for IL property prediction are trained and tested on eleven datasets of different IL properties. The obtained ILTransR presents superior performance as opposed to state-of-the-art models in all benchmarks. The application of ILTransR is exemplified by extensive screening of CO2 absorbent from a huge database of 8 333 096 synthetically-feasible ILs. We are introducing ILTransR, a transfer learning based one-stop framework to predict ionic liquid (IL) properties. High accuracy can be achieved by pre-training the model on millions of unlabeled data and fine-tuning on limited labeled data.