Transfer Learning Facilitates the Prediction of Polymer–Surface Adhesion Strength

Abstract

Machine learning (ML) accelerates the exploration of material properties and their links to the structure of the underlying molecules. In previous work [Shi et al. ACS Applied Materials & Interfaces 2022, 14, 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 data sets 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 data sets by pretraining 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 pretrained model facilitates the prediction accuracy significantly on small data sets. 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.

Calculating Pairwise Similarity of Polymer Ensembles via Earth Mover’s Distance

“…For instance, Aldeghi et al utilized this method to derive a global embedding vector for ensembles of polymer chains for block polymers, random polymers, and alternating copolymers. However, this commonly used average method prematurely reduces the dimensionality of the system, eliminating differences among ensembles due to the topological or monomer sequence information. ,− This premature loss of key information can result in two distinct ensembles being classified as identical. Furthermore, the design of embedding functions becomes nontrivial when the polymer chains have varying chain lengths or complex nonlinear topologies.…”

Empowering Capacitive Devices: Harnessing Transfer Learning for Enhanced Data-Driven Optimization

“…In TL training, given a source domain D s and learning task T s , and a target domain D T and a learning task T T , TL aims to help the learning of the target predictive function f T (·) for the target domain using the knowledge in D s and T s , where D s ≠ D T and T s ≠ T T . The remarkable success of TL has been shown in fields such as materials informatics, − process modeling, − and process monitoring. − In these works, researchers have used different sets of data types including simulated data from empirical or process simulation, , experimental data from related studies, and fake data from generative models to pretrain the ML model before fine-tuning the target problem. Generally, these works showed the positive application of how to use TL to address data limitations affecting the development of accurate data-driven modeling.…”

14 examples of how LLMs can transform materials science and chemistry: a reflection on a large language model hackathon

“…The intersection of machine learning (ML) with chemistry and materials science has witnessed remarkable advancements in recent years. 1–9 Much progress has been made in using ML to, e.g. , accelerate simulations 10,11 or to directly predict properties or compounds for a given application.…”