In this paper, dispersion relations (DRs) of photonic crystals (PhCs) are computed by multilayer perceptron (MLP) and extreme learning machine (ELM) artificial neural networks (ANNs). Bi- and ...tri-dimensional optimized structures presenting distinct DRs and photonic band gaps (PBGs) were selected for case studies. Optical properties of a set of PhCs with similar geometries and different dimensions were calculated by an electromagnetic solver in order to provide input data for ANN training and testing. We demonstrate that simple- and fast-training ANN models are capable of providing accurate DRs' curves in a very short time.
Near-infrared spectroscopy (NIRS) provides broadbands, overtones, and combinations of organic-bond vibrations and has been used to characterize agricultural and food products. The adulteration of ...grated nutmeg with cinnamon is extremely profitable and difficult to detect; to prevent retail fraud, it is vital to differentiate between these materials. This study proposes a model for classifying the adulteration of nutmeg with cinnamon and predicting the level of adulteration. NIR spectra were characterized with six machine learning (ML) algorithms, namely, the principal component-multilayer perceptron (PC-MLP), principal component-linear discriminant analysis (PC-LDA), partial least squares regression (PLSR), support vector machine (SVM), random forest (RF), and decision tree (DT) methods. PC-MLP provided 100% accuracy in calibration and prediction in distinguishing nutmeg from cinnamon. In addition, this approach showed excellent performance in predicting the adulteration ratio of nutmeg and cinnamon with a high coefficient of determination of prediction (R
2
pred
) value of 0.9969, low root mean square error of prediction (RMSEP) value of 0.5728%, and high ratio of prediction to deviation (RPD) value of 17.9605. Therefore, this study indicates the potential of integrating NIR spectroscopy with PC-MLP to classify and quantify the adulteration of nutmeg.
Remote-sensing (RS) images with high spatial and temporal resolutions play a significant role in monitoring periodic landscape changes for earth observation science. To enrich RS images, ...spatiotemporal fusion (STF) is considered a promising approach. The key challenge in the current STF-based methods is the requirement for large-scale data. In this work, we propose a deep-learning-based method called spatiotemporal fusion multilayer perceptron (StfMLP) to tackle this challenge. First, our method focuses on the given data in the manner of transductive learning. Second, we propose a designed multilayer perceptron (MLP) model to capture the time dependency and consistency among the input images. Consequently, StfMLP is capable of simultaneously achieving more accurate fusion and requiring a small-scale of data. We conduct extensive experiments on two widely adopted public datasets, namely Coleambally irrigation area (CIA) and the lower Gwydir catchment (LGC). The experimental results demonstrate that the proposed method outperforms the state-of-the-art methods effectively. Code, trained model, and cropped images are available online ( https://github.com/luhailaing-max/StfMLP-master ).
Gene expression is a key intermediate level that genotypes lead to a particular trait. Gene expression is affected by various factors including genotypes of genetic variants. With an aim of ...delineating the genetic impact on gene expression, we build a deep auto-encoder model to assess how good genetic variants will contribute to gene expression changes. This new deep learning model is a regression-based predictive model based on the MultiLayer Perceptron and Stacked Denoising Auto-encoder (MLP-SAE). The model is trained using a stacked denoising auto-encoder for feature selection and a multilayer perceptron framework for backpropagation. We further improve the model by introducing dropout to prevent overfitting and improve performance.
To demonstrate the usage of this model, we apply MLP-SAE to a real genomic datasets with genotypes and gene expression profiles measured in yeast. Our results show that the MLP-SAE model with dropout outperforms other models including Lasso, Random Forests and the MLP-SAE model without dropout. Using the MLP-SAE model with dropout, we show that gene expression quantifications predicted by the model solely based on genotypes, align well with true gene expression patterns.
We provide a deep auto-encoder model for predicting gene expression from SNP genotypes. This study demonstrates that deep learning is appropriate for tackling another genomic problem, i.e., building predictive models to understand genotypes' contribution to gene expression. With the emerging availability of richer genomic data, we anticipate that deep learning models play a bigger role in modeling and interpreting genomics.
Machine learning is an application of artificial intelligence that focuses on the development of computer algorithms which learn automatically by extracting patterns from the data provided. Machine ...learning techniques can be efficiently used for a problem with a large number of parameters to be optimized and also where it is infeasible to develop an algorithm of specific instructions for performing the task. Here, we combine the finite element simulations and machine learning techniques for the prediction of mode effective indices, power confinement, and coupling length of different integrated photonics devices. Initially, we prepare a dataset using COMSOL Multiphysics and then this data is used for training while optimizing various parameters of the machine learning model. Waveguide width, height, operating wavelength, and other device dimensions are varied to record different modal solution parameters. A detailed study has been carried out for a slot waveguide structure to evaluate different machine learning model parameters including number of layers, number of nodes, choice of activation functions, and others. After training, this model is used to predict the outputs for new input device specifications. This method predicts the output for different device parameters faster than direct numerical simulation techniques. Absolute percentage error of less than 5% in predicting an output has been obtained for slot, strip, and directional waveguide coupler designs. This study paves the step towards using machine learning based optimization techniques for integrated silicon photonics devices.
Visible light positioning (VLP) is a promising indoor localization method as it provides high positioning accuracy and allows for leveraging the existing lighting infrastructure. Photodiode ...(PD)-based receiver is a commonly used tag for VLP. However, a tag employing a single PD requires three or more luminaires to be visible. This article presents a VLP system that uses a custom-made tag utilizing multiple PDs. It applies received signal strength (RSS)-based fingerprinting using a weighted k-nearest neighbor (WkNN) algorithm for localization. Experimental results show that it is possible to localize using less than three luminaires with high accuracy. The Manhattan and Matusita distance metrics are found to provide lower localization accuracy than the Euclidean metric for the WkNN algorithm. The creation of a dense fingerprinting database through 2-D interpolation is presented as a method to reduce the cost of time and labor. The localization performance of the VLP system does not degrade noticeably with the fabricated database. The localization accuracy of the WkNN algorithm is shown to be better than that of a multilayer perceptron (MLP)-based regressor. The developed VLP system is also experimentally benchmarked against the HTC Vive showing comparable performance.
Saturated hydraulic conductivity (K
s
) is an important soil characteristic that controls water moves through the soil. On the other hand, its measurement is difficult, time-consuming, and expensive; ...hence Pedotransfer Functions (PTFs) are commonly used for its estimation. Despite significant development over the years, the PTFs showed poor performance in predicting K
s
. Using Genetic Algorithm (GA), two hybrid Machine Learning based PTFs (ML-PTF), i.e. a combination of GA with Multilayer Perceptron (MLP-GA) and Support Vector Machine (SVM-GA), were proposed in this study. We compared the performances of four machine learning algorithms for different sets of predictors. The predictor combination containing sand, clay, Field Capacity, and Wilting Point showed the highest accuracy for all the ML-PTFs. Among the ML-PTFs, the SVM-GA algorithm outperformed the rest of the PTFs. It was noticed that the SVM-GA PTF demonstrated higher efficiency than the MLP-GA algorithm. The reference model for hydraulic conductivity prediction was selected as the SVM-GA PTF paired with the K-5 predictor variables. The proposed PTFs were compared with 160 models from past literature. It was found that the algorithms advocated were an improvement over these PTFs. The current model would help in efficient spatio-temporal measurement of hydraulic conductivity using pre-available databases.
In this paper, frictionless contact problem for a functionally graded (FG) layer is considered. The FG layer is subjected to load with a rigid punch and the FG layer is bonded on a rigid foundation. ...Analysis of this contact problem was carried out by analytical method, finite element method (FEM) and multilayer perceptron (MLP), comparatively. The main target of this study is to investigate the applicability of MLP analysis for frictionless contact problem of FG layer bonded on a rigid foundation. Analytical solution of the problem is based on the theory of elasticity and integral transform techniques. The physical contact problem is transformed to mathematical system of integral equation. The integral equation in which the contact pressures are unknown functions is numerically solved with the Gauss–Jacobi integration formulation. Finite element analysis of the problem is carried out with ANSYS software by using the two-dimensional modeling technique. Finally, MLP analysis has been used to obtain the contact distances of the problem. Three-layer MLP was used for this calculation. Material properties and loading conditions were created by giving examples of different values in MLP training and testing stages. Program code was rewritten in C++. As a result, average deviation values such as 1.67 and 0.885 were obtained for FEM and MLP, respectively. It has been determined that the contact areas and contact stresses obtained from FEM and MLP are quite compatible with the results obtained from the analytical method.
Objective: This paper is to analyze the performance of the control system of collaborative robots based on cognitive computing technology. Methods: This study combines cognitive computing and deep ...belief network algorithms with collaborative robots to construct a cognitive computing system model based on deep belief networks, which is applied to the control system of collaborative robots. Further, the simulation is used to compare and analyze the algorithm performance of deep belief network (DBN), multilayer perceptron (MLP) and the cognitive computing system model of deep belief network and linear perceptron (DBNLP) proposed in this study. Results: The results show that compared with the DBN and MLP algorithms, the DBNLP algorithm model has a significantly lower error rate in the number of repetitions of the training set, the number of hidden neurons, and the number of network layers. And the number of task backlog, the number of resources to be allocated and the time consumption are less, as well as the accuracy is high. After comparing and analyzing the changes in the estimated value of Ex (expected value), En (entropy value) and He (hyper entropy value), it is found that the estimated value of the DBNLP algorithm model is closer to the true value than that of the DBN and MLP algorithms. Conclusion: The application of the DBNLP algorithm model to collaborative robots can significantly improve its accuracy and safety, providing an experimental basis for the performance improvement of later collaborative robots.
•Analyze the performance of the control system of collaborative robots based on cognitive computing.•This study combines cognitive computing and deep belief network algorithms with collaborative robots to construct a cognitive computing system model based on deep belief networks, which is applied to the control system of collaborative robots.•The simulation is used to compare and analyze the algorithm performance of deep belief network (DBN), multilayer perceptron (MLP) and the cognitive computing system model of deep belief network and linear perceptron (DBNLP) proposed in this study.
Building energy prediction has gained significant attention as a thriving research field owing to its immense potential in enhancing energy efficiency within building energy management systems. ...Therefore, the objective of this study is to predict the values of cooling and heating loads by utilizing the multilayer perceptron neural network for predictive purposes. In this context, a multilayer perceptron neural network is chosen as the core framework for addressing the problem at hand. Subsequently, employing a hybridization approach, multilayer perceptron is combined with eight meta-heuristic algorithms to effectively tune and optimize the hyper-parameters of the multilayer perceptron model. Statistical analysis is conducted to examine the performance of each hybrid model. The findings indicate that MLP-PSOGWO exhibits the best performance, demonstrating the highest levels of accuracy, authenticity, and efficiency. According to the obtained results, it is reported that the MLP-PSOGWO model achieves the highest total R2 values of 0.966 for the cooling load and 0.998 for the heating load. These values surpass those of all other models, indicating that the MLP-PSOGWO model demonstrates the best performance among the hybrid models. Importantly, the results obtained underscore the overall effectiveness of the selected optimizers in delivering accurate outcomes.
Display omitted
•Prediction of cooling and heating loads using MLP neural network and meta-heuristic algorithms.•Eight optimizers are used to effectively tune and optimize the MLP model for accurate energy predictions.•The hybrid MLP-PSOGWO model achieves the highest accuracy and efficiency in load prediction.•Statistical analysis confirms the superior performance of the MLP-PSOGWO model.•MLP-PSOGWO model achieves total R2 values of 0.966 and 0.998 for cooling and heating loads prediction.