Machine learning
(ML) has emerged as one of the most powerful tools
transforming all areas of science and engineering. The nature of molecular
dynamics (MD) simulations, complex and time-consuming calculations,
makes them particularly suitable for ML research. This review article
focuses on recent advancements in developing efficient and accurate
coarse-grained (CG) models using various ML methods, in terms of regulating
the coarse-graining process, constructing adequate descriptors/features,
generating representative training data sets, and optimization of
the loss function. Two classes of the CG models are introduced: bottom-up
and top-down CG methods. To illustrate these methods and demonstrate
the open methodological questions, we survey several important principles
in constructing CG models and how these are incorporated into ML methods
and improved with specific learning techniques. Finally, we discuss
some key aspects of developing machine-learned CG models with high
accuracy and efficiency. Besides, we describe how these aspects are
tackled in state-of-the-art methods and which remain to be addressed
in the near future. We expect that these machine-learned CG models
can address thermodynamic consistent, transferable, and representative
issues in classical CG models.
transparency, excellent resistance to UV radiation, high chemical stability, and reasonably low cost. [1-4] The fabrication of PDMS parts is typically a rapid-prototyping process often achieved by molding or spin-coating a viscous prepolymer with a curing agent. The cross-linking degree depends on the ratio of the prepolymer and the curing agent, which endows PDMS with tunable elastic modulus and stretchability. These characteristics make PDMS one of the most popular stretchable and transparent substrates for biomedical, mechanochromic, and photoelectric devices. [5-7,8] However, the hydrophobic nature of PDMS is a significant obstacle for its combination with hydrophilic materials requiring a strong interfacial adhesion. Various methods have been reported to improve the surface hydrophilicity of PDMS, including surface oxidization via plasma [9-11] or UV-ozone treatment, [12,13] silanization treatment, [14,15] and surface coating with hydrophilic materials. [16,17] Unfortunately, the surface hydrophilicity of PDMS generated by the above methods is typically temporary, which usually only Hydrogels and polydimethylsiloxane (PDMS) are complementary to each other, since the hydrophobic PDMS provides a more stable and rigid substrate, while the water-rich hydrogel possesses remarkable hydrophilicity, biocompatibility, and similarity to biological tissues. Herein a transparent and stretchable covalently bonded PDMS-hydrogel bilayer (PHB) structure is prepared via in situ free radical copolymerization of acrylamide and allylamineexfoliated-ZrP (AA-e-ZrP) on a functionalized PDMS surface. The AA-e-ZrP serves as cross-linking nano-patches in the polymer gel network. The covalently bonded structure is constructed through the addition reaction of vinyl groups of PDMS surface and monomers, obtaining a strong interfacial adhesion between the PDMS and the hydrogel. A mechanical-responsive wrinkle surface, which exhibs transparency change mechanochromism, is created via introducing a cross-linked polyvinyl alcohol film atop the PHB structure. A finite element model is implemented to simulate the wrinkle formation process. The implication of the present finding for the interfacial design of the PHB and PDMS-hydrogel-PVA trilayer (PHPT) structures is discussed.
Phononic bandgap metamaterials, which consist of periodic cellular structures, are capable of absorbing energy within a certain frequency range. Designing metamaterials that trap waves across a wide wave frequency range is still a challenging task. In this paper, we present a deep feature learning-based design framework for both unsupervised generative design and supervised learning-based exploitative optimization. The Gaussian mixture beta variational autoencoder (GM-βVAE) is employed to extract latent features as design variables. Gaussian Process (GP) regression models are trained to predict the relationship between latent features and the properties for the property-driven optimization. The optimal structural designs are reconstructed by mapping the optimized latent feature values to the original image space. Compared with the regular variational autoencoder (VAE), we demonstrate that GM-βVAE has better learning capability and is able to generate a more diversified design set in unsupervised generative design. Furthermore, we propose an iterative GM-βVAE model updating-based design framework. In each iteration, the optimal designs found property-driven optimization are used to update the training dataset. The GM-βVAE model is re-trained with the updated dataset for the optimization search in the next iteration. A comparative study between the traditional single-loop and the iterative approaches is presented to demonstrate the effectiveness of the iterative framework. The caveats to designing phonic bandgap metamaterials are summarized.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.