Transfer Learning Facilitates the Prediction of Polymer-Surface Adhesion Strength

TL;DR

This paper demonstrates that transfer learning with deep neural networks significantly improves polymer-surface adhesion prediction accuracy from small datasets, enabling efficient inverse design across similar systems.

Contribution

It introduces a transfer learning approach for polymer adhesion prediction, reducing data requirements and enhancing model accuracy for related surface systems.

Findings

Transfer learning improves prediction accuracy on small datasets.

Pre-trained models generalize well to similar surfaces.

Optimal network tuning enhances performance.

Abstract

Machine learning (ML) accelerates the exploration of material properties and their links to the structure of the underlying molecules. In previous work [J. Shi, M. J. Quevillon, P. H. A. Valen\c{c}a, and J. K. Whitmer, \textit{ACS Appl. Mater. Interfaces.}, 2022, 14, 32, 37161--37169], ML models were applied to predict the adhesive free energy of polymer--surface interactions with high accuracy from the knowledge of the sequence data, demonstrating successes in inverse-design of polymer sequence for known surface compositions. While the method was shown to be successful in designing polymers for a known surface, extensive datasets were needed for each specific surface in order to train the surrogate models. Ideally, one should be able to infer information about similar surfaces without having to regenerate a full complement of adhesion data for each new case. In the current work, we demonstrate a transfer learning (TL) technique using a deep neural network to improve the accuracy of ML models trained on small datasets by pre-training on a larger database from a related system and fine-tuning the weights of all layers with a small amount of additional data. The shared knowledge from the pre-trained model facilitates the prediction accuracy significantly on small datasets. We also explore the limits of database size on accuracy and the optimal tuning of network architecture and parameters for our learning tasks. While applied to a relatively simple coarse-grained (CG) polymer model, the general lessons of this study apply to detailed modeling studies and the broader problems of inverse materials design.