Deep Neutral Network Research Articles

Deep Neutral Network Enhanced Mesoscopic Thermodynamic Model for Unlocking the Electrode/Electrolyte Interface

Structure and properties of the electrode/electrolyte interface significantly influence the electrochemical processes of energy storage and conversion, yet the challenge lies in accurate description of both molecular characteristics and external field effects. Here, we develop a mesoscopic thermodynamic model that calculates the thermodynamic properties of electrolytes based on chemical potential, and its efficiency is enhanced by a deep neutral network. The deep neutral network enhanced mesoscopic thermodynamic (DeepMT) model effectively bridges the gap between micro‐level characteristics of ions and macro‐level effects of external field, enabling precise presentation of ion density distributions over complex conditions. Our result indicates that the DeepMT model not only demonstrates a computational efficiency improvement of approximately four orders of magnitude over direct theoretical calculations, but also accurately predicts interface properties including ion adsorption, surface charge, and differential capacitance through the statistical analysis of density distributions.

Deep Neutral Network Enhanced Mesoscopic Thermodynamic Model for Unlocking the Electrode/Electrolyte Interface.

Structure and properties of the electrode/electrolyte interface significantly influence the electrochemical processes of energy storage and conversion, yet the challenge lies in accurate description of both molecular characteristics and external field effects. Here, we develop a mesoscopic thermodynamic model that calculates the thermodynamic properties of electrolytes based on chemical potential, and its efficiency is enhanced by a deep neutral network. The deep neutral network enhanced mesoscopic thermodynamic (DeepMT) model effectively bridges the gap between micro-level characteristics of ions and macro-level effects of external field, enabling precise presentation of ion density distributions over complex conditions. Our result indicates that the DeepMT model not only demonstrates a computational efficiency improvement of approximately four orders of magnitude over direct theoretical calculations, but also accurately predicts interface properties including ion adsorption, surface charge, and differential capacitance through the statistical analysis of density distributions.

Integration of deep neutral network modeling and LC-MS-based pseudo-targeted metabolomics to discriminate easily confused ginseng species

Metabolomics covers a wide range of applications in life sciences, biomedicine, and phytology. Data acquisition (to achieve high coverage and efficiency) and analysis (to pursue good classification) are two key segments involved in metabolomics workflows. Various chemometric approaches utilizing either pattern recognition or machine learning have been employed to separate different groups. However, insufficient feature extraction, inappropriate feature selection, overfitting, or underfitting lead to an insufficient capacity to discriminate plants that are often easily confused. Using two ginseng varieties, namely Panax japonicus and P. japonicus var. major, containing the similar ginsenosides, we integrated pseudo-targeted metabolomics and deep neural network (DNN) modeling to achieve accurate species differentiation. A pseudo-targeted metabolomics approach was optimized through data acquisition mode, ion pairs generation, comparison between multiple reaction monitoring (MRM) and scheduled MRM, and chromatographic elution gradient. In total, 1980 ion pairs were monitored within 23 min, allowing for the most comprehensive ginseng metabolome analysis. The established DNN model demonstrated excellent classification performance (in terms of accuracy, precision, recall, F1 score, area under the curve, and receiver operating characteristic) using the entire metabolome data and feature-selection dataset, exhibiting superior advantages over random forest (RF), support vector machine (SVM), extreme gradient boosting (XGBoost), and multilayer perceptron (MLP). Moreover, DNNs were advantageous for automated feature learning, nonlinear modeling, adaptability, and generalization. This study confirmed practicality of the established strategy for efficient metabolomics data analysis and reliable classification performance even when using small-volume samples. This established approach holds promise for plant metabolomics and is not limited to ginseng.

Primary Controlling Factors of Apatite Trace Element Composition and Implications for Exploration in Orogenic Gold Deposits

AbstractSignificant and readily accessible orogenic gold deposits have been previously recognized, exploited, and progressively depleted. Innovative approaches are required to discover new and deeply buried deposits. Recently, trace element variations in apatite have been used to distinguish fertile and barren environments as reliable mineral exploration tools. In this study, machine learning models using Random Forest and Deep Neutral Network are utilized to assess the fertility of quartz veins and altered zones in the orogenic gold systems. The two models have been trained using trace element data of apatite, and the performance of both models yield good classification accuracy (∼90% on average) with low false positive rates. Feature importance analysis (Gini decrease and hidden layer weights) suggest that Pb, As, U, Sr, Eu, Mn, and Fe are the important parameters. Arsenic, U, Eu, Mn, and Fe are redox‐sensitive elements, with their concentrations responding to changes in fluid redox conditions. Strontium primarily originates from the breakdown of plagioclase, which is more likely to occur under oxidizing fluid conditions. Therefore, we infer that the main controlling factor of the models is the redox conditions. The two distinct models consistently highlight the most significant contribution of Pb to this differentiation, even though Pb is not a redox‐sensitive element and can only substitute for Ca2+ in apatite as Pb2+. We infer that the high contribution of Pb may be attributed to the potential transportation of Au in the form of a Pb‐(Bi)‐Au melt, and the Pb content in apatite is influenced by the Pb content in the melt, fluid oxygen, and sulfur fugacity. We also propose a novel discriminant plot using Linear Discriminant Analysis (LDA) to assess the mineralization potential in quartz veins and alteration zones based on apatite trace element data. The machine learning and LDA results suggest that apatite trace elements could be used effectively in the future orogenic gold deposit exploration.

CSC-Net: Cross-Color Spatial Co-Occurrence Matrix Network for Detecting Synthesized Fake Images

Recently, the Generative Adversarial Networks (GAN) generated images have been spread over the social networks, which brings the new challenge in the the community of media forensics. Although some reliable forensic tools have advanced the study of detecting GAN generated images, while the detection accuracy cannot be guaranteed when facing the malicious post-processing attacks, especially in the practical social network scenario. Thus, in this context, we propose a novel well-designed deep neural network equipped with handed-crafted features for dealing with this problem. In particular, relying on the Cross-color Spatial Co-occurrence Matrix (CSCM), the discriminative features are extracted after carefully analyzing and selecting the most effective color channels. Next, the fused features are fed into the deep neutral network for training a high-efficient forensic detector. Extensive experimental results empirically verify that in most detection scenarios, our proposed detector performs superiorly to the prior arts, especially in the case of post-processing attacks. Moreover, we also highlight the relevance of the proposed detector over the realistic social network platforms, and its generalization capability in three different scenarios.

Fractional Fourier transform imaging based on untrained neural networks

Fractional Fourier transform is an important branch of optical research, and it is widely used in optical encryption, optical filtering, image watermarking and other fields. The phase retrieval in the case of fractional Fourier transform is widely studied. Also, deep learning has been an intriguing method for optical computational imaging. However, in optical computational imaging, traditional deep learning methods possess some intrinsic disadvantages. In optical imaging experiments, it is often difficult to obtain sufficient quality and quantity of labeled data for training, thus leading to poor robustness of the trained neural network. Even with sufficient datasets, the training time can be particularly long. In recent years, there has been an increase in interest in physic-driven untrained neural networks for computational imaging. Herein we use such a method to study the fractional Fourier transform imaging, which combines neural networks with optical models to achieve phase retrieval of fractional Fourier transform. Unlike the traditional neural network training with the original image as the target, our network framework is used only a single intensity image for the phase retrieval of fractional Fourier transform images. The output image of the neural network will serve as an optical model through fractional Fourier transform, and then the output image of the optical model will be used as a loss function to drive the neural network training with the output image of the neural network. We study the fractional Fourier transform reconstruction for the cases where the fractional order is less than 1 and greater than 1. The simulations and experiments show that the network framework can implement the fractional Fourier transform reconstructions of the intensity objects and phase objects for different fraction orders, in which only 2000 iterations are needed. The experimental results show that the similarity between the reconstructed image and the original image, i.e. the number of normalized correlation coefficient, can reach 99.7%. Therefore, our work offers an efficient scheme for functional Fourier transform reconstruction with physics-enhanced deep neutral network.

A Deep Learning Feature Fusion Based Health Index Construction Method for Prognostics Using Multiobjective Optimization

Degradation modeling and prognostics serve as the basis for system health management. Recently, various sensors provide plentiful monitoring data that can reflect the system status. A multitude of feature fusion techniques based on multisensor data have been proposed to generate a composite health index (HI) for prognostics, which can represent the underlying degradation mechanism. Most existing methods have used linear fusion models and neglected the practical requirements for HI construction, which are insufficient to reveal the nonlinear relations among features and difficult to obtain accurate HIs for complicated systems. This study proposes a novel feature fusion-based HI construction method with deep learning and multiobjective optimization. Multiple degradation features are fused by a deep neutral network (DNN). Several desired properties that the HIs should have for prognostics are adopted to formulate the objective functions of DNN training. To balance the spatial complexity and performance of the fusion model, a multiobjective optimization model is generated for training the DNN. Then, a generalized nonlinear Wiener process model is used to predict the remaining useful life with the resulted HIs. Finally, two cases are analyzed to illustrate the effectiveness and robustness of the proposed method.

AN IMPROVED ACCURACY FOR THE FORECASTING OF POWER GENERATION OVER A LONG-TERM HORIZON

Renewable energy becomes increasingly popular in the global electric energy grid, improving the accuracy of renewable energy forecasting is critical to power system planning, management, and operations. However, this is a challenging task due to the intermittent and chaotic nature of renewable energy data. To date, various methods have been developed, including physical models, statistical methods, artificial intelligence techniques, and their hybrids to improve the forecasting accuracy of renewable energy. Hence this research proposed to hybridize two strong deep learning algorithms where modeling of more complex functioning is allowed by the use of multiple layers of abstraction in order to come up with a powerful forecasting model to predict solar power generation over long term horizon. Finally, the Deep Neutral Network and Long-short Term memory Network (DNN-LSTM) method can generate predicted solar energy consumption in a fully connected hierarchy. The proposed DNN-LSTM model achieved Mean Square Error (MSE) of 0.00825 and MAE of 0.00100 respectively. This is by far the lowest value when compare against the existing model i.e MLSHM which has MSE of 0.05700 and MAPE of 0.00695, LSTM which has MSE of 0.0536 and MAE of 0.0037 and Gated Recurrent Unit (GRU) which has MSE of 0.03460 and MAE of 0.00243 respectively. Thus, the proposed DNN-LSTM have clearly enhanced the forecasting accuracy as against all the existing models that was used for the evaluation and achieved the lowest values in terms of validation of MSE and MAE.

An Enhanced Deep Neural Network-Based Approach for Speaker Recognition Using Triumvirate Euphemism Strategy

Automatic Speech Recognition (ASR) has been an intensive research area during the recent years in internet to enable natural human–machine communication. However, the existing Deep Neutral Network (DNN) techniques need more focus on feature extraction process and recognition accuracy. Thus, an enhanced deep neural network (DNN)-based approach for speaker recognition with a novel Triumvirate Euphemism Strategy (TES) is proposed. This overcomes poor feature extraction from Mel-Frequency Cepstral Coefficient (MFCC) map by extracting the features based on petite, hefty and artistry of the features. Then, the features are trained with Silhouette Martyrs Method (SMM) without any inter-class and intra-class separability problems and margins are affixed between classes with three new loss functions, namely A-Loss, AM-Loss and AAM-Loss. Additionally, the parallelization is done by a mini-batch-based BP algorithm in DNN. A novel Frenzied Heap Atrophy (FHA) with a multi-GPU model is introduced in addition with DNN to enhance the parallelized computing that accelerates the training procedures. Thus, the outcome of the proposed technique is highly efficient that provides feasible extraction features and gives incredibly precise results with 97.5% accuracy in the recognition of speakers. Moreover, various parameters were discussed to prove the efficiency of the system and also the proposed method outperformed the existing methods in all aspects.

A hybrid intelligent genetic algorithm for truss optimization based on deep neutral network

The truss optimization problem has been extensively investigated, and the optimized trusses have been widely used in various fields. Truss optimization is a challenging optimization involving many difficulties, such as discrete variables and non-convex problems. In order to solve these problems, various metaheuristic optimization methods have been proposed. Due to the stochastic feature of these methods, it is always computationally intensive, and the optimization results may vary greatly in different optimization runs. In this paper, a hybrid intelligent genetic algorithm (HIGA) is proposed to improve the effectiveness and efficiency of truss optimization problems. This method systematically integrates deep neural network (DNN) and genetic algorithm (GA) in the optimization process. A two-step training procedure is proposed where the data generated during the optimization process is exploited to update DNN. Next, based on the generated DNN model, an optimization prediction procedure is proposed to seek more optimized trusses in an efficient manner. Through the investigation of three classical truss problems (size optimization, shape optimization and size-shape integrated optimization), the effectiveness of the proposed method is validated. In addition, the influence of different settings on the optimization performance and efficiency is investigated to demonstrate the applicability and robustness of the proposed method.

Supervised Machine Learning Algorithms for Predicting Rate Constants of Ozone Reaction with Micropollutants

The second-order rate constants of organic contaminants degraded by ozone (kO3) are of great importance for evaluating their treatment efficiency and optimizing treatment processes. In this work, several supervised machine learning (ML) algorithms, including multiple linear regression (MLR), support vector machine with radial basis function kernels (SVM-RBF), decision tree (DT), random forest (RF), and deep neutral network (DNN) methods, were used to develop quantitative structure–property relationship (QSPR) models for the estimation of log kO3. What is more, a series of quantum chemical and newly proposed norm descriptors was successfully used in developing ML models as inputs. The statistical parameters correlation coefficient (R2), mean square error (MSE), mean absolute error (MAE), and external validation parameter (Qext2) were used to evaluate the accuracy, robustness, and predictability of the as-developed models, suggesting that the nonlinear models (especially for the RF model) have better performance in predicting log kO3 values than the linear model. It is expected that the proposed norm descriptors can be employed to evaluate other reaction rate constants or chemical properties.

Intelligent Control on Urban Natural Gas Supply Using a Deep-Learning-Assisted Pipeline Dispatch Technique

Natural gas has been attracting increasing attentions all around the world as a relatively cleaner energy resource compared with coal and crude oil. Except for the direct consumption as fuel, electricity generation is now another environmentally-friendly utilization of natural gas, which makes it more favorable as the energy supply for urban areas. Pipeline transportation is the main approach connecting the natural gas production field and urban areas thanks to the safety and economic reasons. In this paper, an intelligent pipeline dispatch technique is proposed using deep learning methods to predict the change of energy supply to the urban areas as a consequence of compressor operations. Practical operation data is collected and prepared for the training and validation of deep learning models, and the accelerated predictions can help make controlling plans regarding compressor operations to meet the requirement in urban natural gas supply. The proposed deep neutral network is equipped with self-adaptability, which enables the general adaption on various temporal compressor conditions including failure and maintenance.

Learning physics at future e\u2212e+ colliders with machine

Information deformation and loss in jet clustering are one of the major limitations for precisely measuring hadronic events at future e−e+ colliders. Because of their dominance in data, the measurements of such events are crucial for advancing the precision frontier of Higgs and electroweak physics in the next decades. We show that this difficulty can be well-addressed by synergizing the event-level information into the data analysis, with the techniques of deep neutral network. In relation to this, we introduce a CMB-like observable scheme, where the event-level kinematics is encoded as Fox-Wolfram (FW) moments at leading order and multi-spectra of spherical harmonics at higher orders. Then we develop a series of jet-level (w/ and w/o the FW moments) and event-level classifiers, and analyze their sensitivity performance comparatively with two-jet and four-jet events. As an application, we analyze measuring Higgs decay width at e−e+ colliders with the data of 5ab−1@240GeV. The precision obtained is significantly better than the baseline ones presented in documents. We expect this strategy to be applied to many other hadronic- event measurements at future e−e+ colliders, and to open a new angle for evaluating their physics capability.

Locate the Superficial Femoral Artery with Occlusion by Deep Neural Network Correcting Interpolation.

In clinical practice, doctors usually use computed tomography angiography (CTA) to examine lower extremity atherosclerotic occlusive (ASO). Conveniently and accurately locating occlusive superficial femoral artery (SFA) which is difficult to extract from CTA can facilitate diagnosis and surgery. This paper proposed a method locating the occlusive SFA from CTA conveniently. The proposed method first takes control points at a certain interval to bicubic interpolate, and then feeds image patches generated based on the interpolation results to deep neutral network (DNN) to obtain vessel center points. The final location error is less than 9 pixels, which meets the requirements of clinical assessment accuracy. It can be used to assist the diagnosis and surgery of ASO.

Deep Learning-Enhanced Framework for Performance Evaluation of a Recommending Interface with Varied Recommendation Position and Intensity Based on Eye-Tracking Equipment Data Processing

The increasing amount of marketing content in e-commerce websites results in the limited attention of users. For recommender systems, the way recommended items are presented becomes as important as the underlying algorithms for product selection. In order to improve the effectiveness of content presentation, marketing experts experiment with the layout and other visual aspects of website elements to find the most suitable solution. This study investigates those aspects for a recommending interface. We propose a framework for performance evaluation of a recommending interface, which takes into consideration individual user characteristics and goals. At the heart of the proposed solution is a deep neutral network trained to predict the efficiency a particular recommendation presented in a selected position and with a chosen degree of intensity. The proposed Performance Evaluation of a Recommending Interface (PERI) framework can be used to automate an optimal recommending interface adjustment according to the characteristics of the user and their goals. The experimental results from the study are based on research-grade measurement electronics equipment Gazepoint GP3 eye-tracker data, together with synthetic data that were used to perform pre-assessment training of the neural network.

Multi-task deep neural network (MT-DNN) enabled optical performance monitoring from directly detected PDM-QAM signals.

We experimentally demonstrate a multi-task deep neutral network (MT-DNN) enabled optical performance monitoring (OPM) for PDM-QPSK/8QAM/16QAM signals, by using the asynchronous amplitude histogram (AAH) after the direct detection. Consequently, we can simultaneously realize the modulation format identification (MFI), baud rate identification (BRI), and optical signal-to-noise ratio (OSNR) monitoring simultaneously. In the simulation, when both 20Gbaud and 30Gbaud PDM-QPSK, PDM-8QAM and PDM-16QAM signals are taken into account, the accuracies of both MFI and BRI are 100%. Meanwhile, the root-mean-square error (RMSE) of OSNR monitoring is 0.58dB over a range of 10-22dB, 14-24dB, and 17-26dB for PDM-QPSK, PDM-8QAM and PDM-16QAM, respectively. Furthermore, the numerical results show that RMSE of OSNR monitoring is degraded from 0.58dB to 0.97dB, when chromatic dispersion (CD) is accumulated from 0 to 1600ps/nm. Meanwhile, the MFI accuracy is degraded from 100% to 97.25%, and the BRI accuracy remains 100%. When 2.8Gbaud and 9.8Gbaud signals are used for the experimental verification under the condition of back-to-back transmission, the accuracies of MFI and BRI are 100% and 96.8%, respectively, and the RMSE of OSNR monitoring is 0.76dB. Since only one photodetector (PD) with the asynchronous sampling and one MT-DNN are required, the proposed OPM scheme has the advantage of high cost performance.

Identifying Mg ii narrow absorption lines with deep learning

Metal absorption line systems in distant quasar spectra probe of the history of gas content in the universe. The MgII $\lambda \lambda$ 2796, 2803 doublet is one of the most important absorption lines since it is a proxy of the star formation rate and a tracer of the cold gas associated with high redshift galaxies. Machine learning algorithms have been used to detect absorption lines systems in large sky surveys, such as Principle Component Analysis (PCA), Gaussian Process (GP) and decision trees. A very powerful algorithm in the field of machine learning called deep neural networks, or '' deep learning'' is a new structure of neural network that automatically extracts semantic features from raw data and represents them at a high level. In this paper, we apply a deep convolutional neural network for absorption line detection. We use the previously published DR7 MgII catalog (Zhu et al. 2013) as the training and validation sample and the DR12 MgII catalog as the test set. Our deep learning algorithm is capable of detecting MgII absorption lines with an accuracy of $\sim$94% . It takes only $\sim 9$ seconds to analyze $\sim$ 50000 quasar spectra with our deep neural network, which is ten thousand times faster than traditional methods, while preserving high accuracy with little human interference. Our study shows that Mg II absorption line detection accuracy of a deep neutral network model strongly depends on the filter size in the filter layer of the neural network, and the best results are obtained when the filter size closely matches the absorption feature size.

Biomedical Named Entity Recognition based on Deep Neutral Network

Biomedical Named Entity Recognition based on Deep Neutral Network