Comparison of Different Neural Network Architectures for Plasmonic Inverse Design.
Ontology highlight
ABSTRACT:
SUBMITTER: Wu Q
PROVIDER: S-EPMC8444196 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Predicting physical response of an artificially structured material is of particular interest for scientific and engineering applications. Here we use deep learning to predict optical response of artificially engineered nanophotonic devices. In addition to predicting forward approximation of transmission response for any given topology, this approach allows us to inversely approximate designs for a targeted optical response. Our Deep Neural Network (DNN) could design compact (2.6 × 2.6 μm2) silicon-on-insulator (SOI)-based 1 × 2 power splitters with various target splitting ratios in a fraction of a second. This model is trained to minimize the reflection (to smaller than ~ -20 dB) while achieving maximum transmission efficiency above 90% and target splitting specifications. This approach paves the way for rapid design of integrated photonic components relying on complex nanostructures.
| S-EPMC6361971 | biostudies-literature
Project description:ObjectiveAttention networks learn an intelligent weighted averaging mechanism over a series of entities, providing increases to both performance and interpretability. In this article, we propose a novel time-aware transformer-based network and compare it to another leading model with similar characteristics. We also decompose model performance along several critical axes and examine which features contribute most to our model's performance.Materials and methodsUsing data sets representing patient records obtained between 2017 and 2019 by the Kaiser Permanente Mid-Atlantic States medical system, we construct four attentional models with varying levels of complexity on two targets (patient mortality and hospitalization). We examine how incorporating transfer learning and demographic features contribute to model success. We also test the performance of a model proposed in recent medical modeling literature. We compare these models with out-of-sample data using the area under the receiver-operator characteristic (AUROC) curve and average precision as measures of performance. We also analyze the attentional weights assigned by these models to patient diagnoses.ResultsWe found that our model significantly outperformed the alternative on a mortality prediction task (91.96% AUROC against 73.82% AUROC). Our model also outperformed on the hospitalization task, although the models were significantly more competitive in that space (82.41% AUROC against 80.33% AUROC). Furthermore, we found that demographic features and transfer learning features which are frequently omitted from new models proposed in the EMR modeling space contributed significantly to the success of our model.DiscussionWe proposed an original construction of deep learning electronic medical record models which achieved very strong performance. We found that our unique model construction outperformed on several tasks in comparison to a leading literature alternative, even when input data was held constant between them. We obtained further improvements by incorporating several methods that are frequently overlooked in new model proposals, suggesting that it will be useful to explore these options further in the future.
| S-EPMC8358476 | biostudies-literature
Project description:In the recent decade, computational tools have become central in material design, allowing rapid development cycles at reduced costs. Machine learning tools are especially on the rise in photonics. However, the inversion of the Maxwell equations needed for the design is particularly challenging from an optimization standpoint, requiring sophisticated software. We present an innovative, open-source software tool called Neural Inverse Design of Nanostructures (NIDN) that allows designing complex, stacked material nanostructures using a physics-based deep learning approach. Instead of a derivative-free or data-driven optimization or learning method, we perform a gradient-based neural network training where we directly optimize the material and its structure based on its spectral characteristics. NIDN supports two different solvers, rigorous coupled-wave analysis and a finite-difference time-domain method. The utility and validity of NIDN are demonstrated on several synthetic examples as well as the design of a 1550 nm filter and anti-reflection coating. Results match experimental baselines, other simulation tools, and the desired spectral characteristics. Given its full modularity in regard to network architectures and Maxwell solvers as well as open-source, permissive availability, NIDN will be able to support computational material design processes in a broad range of applications.
| S-EPMC9780235 | biostudies-literature
Project description:Noniridescent and nonfading structural colors generated from metallic and dielectric nanoparticles with extraordinary optical properties hold great promise in applications such as image display, color printing, and information security. Yet, due to the strong wavelength dependence of optical constants and the radiation pattern, it is difficult and time-consuming to design nanoparticles with the desired hue, saturation, and brightness. Herein, we combined the Monte Carlo and Mie scattering simulations and a bidirectional neural network (BNN) to improve the design of gold nanoparticles' structural colors. The optical simulations provided a dataset including color properties and geometric parameters of gold nanoparticle systems, while the BNN was proposed to accurately predict the structural colors of gold nanoparticle systems and inversely design the geometric parameters for the desired colors. Taking the human chromatic discrimination ability as a criterion, our proposed approach achieved a high accuracy of 99.83% on the predicted colors and 98.5% on the designed geometric parameters. This work provides a general method to accurately and efficiently design the structural colors of nanoparticle systems, which can be exploited in a variety of applications and contribute to the development of advanced optical materials.
| S-EPMC8703294 | biostudies-literature
Project description:Modern artificial neural networks (ANNs) have long been designed on foundations of mathematics as opposed to their original foundations of biomimicry. However, the structure and function of these modern ANNs are often analogous to real-life biological networks. We propose that the ubiquitous information-theoretic principles underlying the development of ANNs are similar to the principles guiding the macro-evolution of biological networks and that insights gained from one field can be applied to the other. We generate hypotheses on the bow-tie network structure of the Janus kinase - signal transducers and activators of transcription (JAK-STAT) pathway, additionally informed by the evolutionary considerations, and carry out ANN simulation experiments to demonstrate that an increase in the network's input and output complexity does not necessarily require a more complex intermediate layer. This observation should guide novel biomarker discovery-namely, to prioritize sections of the biological networks in which information is most compressed as opposed to biomarkers representing the periphery of the network.
| S-EPMC9929672 | biostudies-literature
Project description:This data manuscript presents a set of signals collected from the Llaima volcano located at the western edge of the Andes in Araucania Region, Chile. The signals were recorded from the LAV station between 2010 and 2016. After individually processing and analyzing every signal, specialists from the Observatorio Vulcanológico de los Andes Sur (OVDAS) classified them into four class according to their event source: i) Volcano-Tectonic (VT); ii) Long Period (LP); iii) Tremor (TR), and iv) Tectonic (TC). The dataset is composed of 3592 signals separated by class and filtered to select the segment that contains the most representative part of the seismic event. This dataset is important to support researchers interested in studying seismic signals from active volcanoes and developing new methods to model time-dependent data. In this sense, we have published the manuscript "In-Depth Comparison of Deep Artificial Neural Network Architectures on Seismic Events Classification" [1] analyzing such signals with different Deep Neural Networks (DNN). The main contribution of such manuscript is a new DNN architecture called SeismicNet, which provided classification results among the best in the literature without demanding explicit signal pre-processing steps. Therefore, the reader is referred to such manuscript for the interpretation of the data.
| S-EPMC7206203 | biostudies-literature
Project description:MotivationConvolutional neural networks (CNN) have outperformed conventional methods in modeling the sequence specificity of DNA-protein binding. Yet inappropriate CNN architectures can yield poorer performance than simpler models. Thus an in-depth understanding of how to match CNN architecture to a given task is needed to fully harness the power of CNNs for computational biology applications.ResultsWe present a systematic exploration of CNN architectures for predicting DNA sequence binding using a large compendium of transcription factor datasets. We identify the best-performing architectures by varying CNN width, depth and pooling designs. We find that adding convolutional kernels to a network is important for motif-based tasks. We show the benefits of CNNs in learning rich higher-order sequence features, such as secondary motifs and local sequence context, by comparing network performance on multiple modeling tasks ranging in difficulty. We also demonstrate how careful construction of sequence benchmark datasets, using approaches that control potentially confounding effects like positional or motif strength bias, is critical in making fair comparisons between competing methods. We explore how to establish the sufficiency of training data for these learning tasks, and we have created a flexible cloud-based framework that permits the rapid exploration of alternative neural network architectures for problems in computational biology.Availability and implementationAll the models analyzed are available at http://cnn.csail.mit.eduContactgifford@mit.eduSupplementary informationSupplementary data are available at Bioinformatics online.
| S-EPMC4908339 | biostudies-literature
Project description:Generating optimal nanomaterials using artificial neural networks can potentially lead to a notable revolution in future materials design. Although progress has been made in creating small and simple molecules, complex materials such as crystalline porous materials have yet to be generated using any of the neural networks. Here, we have implemented a generative adversarial network that uses a training set of 31,713 known zeolites to produce 121 crystalline porous materials. Our neural network takes in inputs in the form of energy and material dimensions, and we show that zeolites with a user-desired range of 4 kJ/mol methane heat of adsorption can be reliably produced using our neural network. The fine-tuning of user-desired capability can potentially accelerate materials development as it demonstrates a successful case of inverse design of porous materials.
| S-EPMC6941911 | biostudies-literature
Project description:BackgroundRecent years have seen a surge of novel neural network architectures for the integration of multi-omics data for prediction. Most of the architectures include either encoders alone or encoders and decoders, i.e., autoencoders of various sorts, to transform multi-omics data into latent representations. One important parameter is the depth of integration: the point at which the latent representations are computed or merged, which can be either early, intermediate, or late. The literature on integration methods is growing steadily, however, close to nothing is known about the relative performance of these methods under fair experimental conditions and under consideration of different use cases.ResultsWe developed a comparison framework that trains and optimizes multi-omics integration methods under equal conditions. We incorporated early integration, PCA and four recently published deep learning methods: MOLI, Super.FELT, OmiEmbed, and MOMA. Further, we devised a novel method, Omics Stacking, that combines the advantages of intermediate and late integration. Experiments were conducted on a public drug response data set with multiple omics data (somatic point mutations, somatic copy number profiles and gene expression profiles) that was obtained from cell lines, patient-derived xenografts, and patient samples. Our experiments confirmed that early integration has the lowest predictive performance. Overall, architectures that integrate triplet loss achieved the best results. Statistical differences can, overall, rarely be observed, however, in terms of the average ranks of methods, Super.FELT is consistently performing best in a cross-validation setting and Omics Stacking best in an external test set setting.ConclusionsWe recommend researchers to follow fair comparison protocols, as suggested in the paper. When faced with a new data set, Super.FELT is a good option in the cross-validation setting as well as Omics Stacking in the external test set setting. Statistical significances are hardly observable, despite trends in the algorithms' rankings. Future work on refined methods for transfer learning tailored for this domain may improve the situation for external test sets. The source code of all experiments is available under https://github.com/kramerlab/Multi-Omics_analysis.
| S-EPMC9926634 | biostudies-literature
Project description: Not available
| S-EPMC10545200 | biostudies-literature