Abstract:In the present work, we address the problem of utilizing machine learning (ML) methods to predict the thermal properties of polymers by establishing "structure-property" relationships. Having focused on a particular class of heterocyclic polymers, namely polyimides (PIs), we developed a graph convolutional neural network (GCNN), being one of the most promising tools for working with big data, to predict the PI glass transition temperature T g as an example of the fundamental property of polymers. To train the GCNN, we propose an original methodology based on using a "transfer learning" approach with an enormous "synthetic" data set for pretraining and a small experimental data set for its fine-tuning. The "synthetic" data set contains more than 6 million combinatorically generated repeating units of PIs and theoretical values of their T g values calculated using the well-established Askadskii's quantitative structure-property relationship (QSPR) computational scheme. Additionally, an experimental data set for 214 PIs was also collected from the literature for training, fine-tuning, and validation of the GCNN. Both "synthetic" and experimental data sets are included into a PolyAskInG database (Polymer Askadskii's Intelligent Gateway). By using the PolyAskInG database, we developed GCNN which allows estimation of T g of PI with a mean absolute error (MAE) of about 20 K, which is 1.5 times lower than in the case of Askadskii QSPR analysis (33 K). To prove the efficiency and usability of the proposed GCNN architecture and training methodology for predicting polymer properties, we also employed "transfer learning" to develop alternative GCNN pretrained on proxy-characteristics taken from the popular quantum-chemical QM9 database for small compounds and fine-tuned on an experimental T g values data set from PolyAskInG database. The obtained results indicate that pretraining of GCNN on the "synthetic" polymer data set provides MAE which is almost twice as low as that in the case of using the QM9 data set in the pretraining stage (∼41 K). Furthermore, we address the questions associated with the influence of the differences in the size of the experimental and "synthetic" data sets (so-called "reality gap" problem), as well as their chemical composition on the training quality. Our results state the overall priority of using polymer data sets for developing deep neural networks, and GCNN in particular, for efficient prediction of polymer properties. Moreover, our work opens up a challenge for the theoretically supported generation of large "synthetic" data sets of polymer properties for the training of the complex ML models. The proposed methodology is rather versatile and may be generalized for predicting other properties of different polymers and copolymers synthesized through the polycondensation reaction.

Large-Scale Glass-Transition Temperature Prediction with an Equivariant Neural Network for Screening Polymers

Efficient Prediction of Polymer Glass Transition Temperatures through Machine Learning Methods

Machine learning analysis of a large set of homopolymers to predict glass transition temperatures

Predicting Polymers’ Glass Transition Temperature by a Chemical Language Processing Model

A neural network approach to prediction of glass transition temperature of polymers

Benchmarking Machine Learning Models for Polymer Informatics: An Example of Glass Transition Temperature

Machine Learning Prediction of Glass Transition Temperature of Conjugated Polymers from Chemical Structure

Glass Transition Temperature Prediction of Polymers via Graph Reinforcement Learning

Molecular Dynamics Simulation on the Glass Transition Temperatures of TPI/NR/LDPE

Machine Learning with Enormous "Synthetic" Data Sets: Predicting Glass Transition Temperature of Polyimides Using Graph Convolutional Neural Networks

Prediction of Glass Transition Temperature of Polymers Using Simple Machine Learning

Prediction and Interpretability of Glass Transition Temperature of Homopolymers by Data-Augmented Graph Convolutional Neural Networks

Polymer design using machine learning: a quest for high glass transition temperature

High Glass Transition Temperature Fluorinated Polymers Based on Transfer Learning with Small Experimental Data

Predicting glass transition of amorphous polymers by application of cheminformatics and molecular dynamics simulations

Assessing and Improving Machine Learning Model Predictions of Polymer Glass Transition Temperatures

Interpretable Machine Learning Framework to Predict the Glass Transition Temperature of Polymers

A combinatorial machine-learning-driven approach for predicting glass transition temperature based on numerous molecular descriptors

Predicting the Glass Transition Temperature of Biopolymers via High-Throughput Molecular Dynamics Simulations and Machine Learning

A glimpse inside materials: Polymer structure – Glass transition temperature relationship as observed by a trained artificial intelligence

A fast transferable method for predicting the glass transition temperature of polymers from chemical structure