In the field of marine detection and warning, predicting the heights of ocean wave is a very important project. In order to predict the ocean wave heights accurately and quickly, our methodology ...utilizes a hybrid Mind Evolutionary Algorithm-BP neural network strategy (MEA-BP). This paper investigates how the BP neural network (BPnn) evolution with MEA improves the generalization ability and predictability of BPnn. The MEA-BP model combines the local searching ability of the BPnn and the global searching ability of the MEA which can avoid premature convergence and poor prediction effect. In order to search individuals which contain optimal weights and thresholds, the MEA searches all the initial weights and thresholds intelligently by similartaxis and dissimilation operation, finally assign them to the initial BPnn. The study is conducted using data collected from 12 observation points across two geographically distinct regions, Bohai Sea, Yellow Sea, for the period from Jan 1, 2016 to Dec 31, 2016. The data is chosen such that the study covers a wide range of geographical locations and different weather. We compare the prediction performance and generalization capabilities of MEA-BP with the Genetic Algorithm-BP neural network model (GA-BP) which also developed with the BPnn. The performance study results demonstrate that MEA-BP performs better than the GA-BP and Standard BP neural network model (St-BP) with faster running time and higher prediction accuracy.
•MEA can remember more than one generation of evolutionary information.•Similartaxis and dissimilation operations towards a correct direction, never become worse.•Similartaxis and dissimilation operations are easier to find the optimal individual, saving the number of iterations.•Crossover and mutation operations in GA can produce good genes, or destroy the original genes.•The study data covers a wide range of geographical locations and different weather to eliminate the effects of geography and weather.
This paper proposes an innovative global solution which is a pioneering work applying automated machine learning algorithms to remarkable precision sparse underwater direction-of-arrival (DOA) ...estimation that views the subaquatic sparse-sampling DOA estimation problem as a classification prediction task. The proposed solution, termed automated multi-layer perceptron discriminative neural network (AutoMPDNN), is built upon a Bayesian optimization framework. AutoMPDNN transforms sparsely sampled time-domain signals into the complex domain, preserving essential components in a one-source single-snapshot scenario. Leveraging Bayesian optimization principles, the algorithm embeds necessary hyperparameters into the loss function, effectively defining it as a maximum likelihood problem using the upper confidence bound function and incorporating sparse signal features. We also explore the model space architecture and introduce variants of AutoMPDNN, denoted as AutoMPDNNs_ln (n = 2,3,4). Through a series of plane wave simulation experiments, it is demonstrated that AutoMPDNN achieves the highest prediction performance for one-source single-snapshot scenarios compared to classical DOA estimation algorithms that incorporate sparse representation approaches, as well as contemporary deep learning DOA methods under varying conditions.
Conclusion
In this paper, an iterative framework
iterUA
is designed for publishing frequency distribution with reduced relative error on multidimensional data under LDP. In each iteration step, the ...optimized user allocation strategy OUAS is invoked to reduce the relative error in the derived results. In the furture, we are going to investigate the problem of the relative error optimization for mean estimation under LDP.
Due to the lack of a specific design for scenarios such as scale change, illumination difference, and occlusion, current person re-identification methods are difficult to put into practice. A ...Multi-Branch Feature Fusion Network (MFFNet) is proposed, and Shallow Feature Extraction (SFF) and Multi-scale Feature Fusion (MFF) are utilized to obtain robust global feature representations while leveraging the Hybrid Attention Module (HAM) and Anti-erasure Federated Block Network (AFBN) to solve the problems of scale change, illumination difference and occlusion in scenes. Finally, multiple loss functions are used to efficiently converge the model parameters and enhance the information interaction between the branches. The experimental results show that our method achieves significant improvements over Market-1501, DukeMTMC-reID, and MSMT17. Especially on the MSMT17 dataset, which is close to real-world scenarios, MFFNet improves by 1.3 and 1.8% on Rank-1 and mAP, respectively.
As the most widely used storage device today, hard disks are efficient and convenient, but the damage incurred in the event of a failure can be very significant. Therefore, early warnings before hard ...disk failure, allowing the stored content to be backed up and transferred in advance, can reduce many losses. In recent years, an endless stream of research on the prediction of hard disk failure prediction has emerged. The detection accuracy of various methods, from basic machine learning models, such as decision trees and random forests, to deep learning methods, such as BP neural networks and recurrent neural networks, has also been improving. In this paper, based on the idea of blending ensemble learning, a novel failure prediction method combining machine learning algorithms and neural networks is proposed on the publicly available BackBlaze hard disk datasets. The failure prediction experiment is conducted only with S.M.A.R.T., that is, the learned characteristics collected by self-monitoring analysis and reporting technology, which are internally counted during the operation of the hard disk. The experimental results show that this ensemble learning model is able to outperform other independent models in terms of evaluation criterion based on the Matthews correlation coefficient. Additionally, through the experimental results on multiple types of hard disks, an ensemble learning model with high performance on most types of hard disks is found, which solves the problem of the low robustness and generalization of traditional machine learning methods and proves the effectiveness and high universality of this method.
Early risk prediction of diabetes could help doctors and patients to pay attention to the disease and intervene as soon as possible, which can effectively reduce the risk of complications. In this ...paper, a GA-stacking ensemble learning model is proposed to improve the accuracy of diabetes risk prediction. Firstly, genetic algorithms (GA) based on Decision Tree (DT) is used to select individuals with high adaptability, that is, a subset of attributes suitable for diabetes risk prediction. Secondly, the optimized convolutional neural network (CNN) and support vector machine (SVM) are used as the primary learners of stacking to learn attribute subsets, respectively. Then, the output of CNN and SVM is used as the input of the mate learner, the fully connected layer, for classification. Qingdao desensitization physical examination data from 1 January 2017 to 31 December 2019 is used, which includes body temperature, BMI, waist circumference, and other indicators that may be related to early diabetes. We compared the performance of GA-stacking with K-nearest neighbor (KNN), SVM, logistic regression (LR), Naive Bayes (NB), and CNN before and after adding GA through the average prediction time, accuracy, precision, sensitivity, specificity, and F1-score. Results show that prediction efficiency can be improved by adding GA. GA-stacking has higher prediction accuracy. Moreover, the strong generalization ability and high prediction efficiency of GA-stacking have also been verified on the early-stage diabetes risk prediction dataset published by UCI.
With the rapid development of deep learning, researchers are actively exploring its applications in the field of industrial anomaly detection. Deep learning methods differ significantly from ...traditional mathematical modeling approaches, eliminating the need for intricate mathematical derivations and offering greater flexibility. Deep learning technologies have demonstrated outstanding performance in anomaly detection problems and gained widespread recognition. However, when dealing with multivariate data anomaly detection problems, deep learning faces challenges such as large-scale data annotation and handling relationships between complex data variables. To address these challenges, this study proposes an innovative and lightweight deep learning model—the Attention-Based Deep Convolutional Autoencoding Prediction Network (AT-DCAEP). The model consists of a characterization network based on convolutional autoencoders and a prediction network based on attention mechanisms. The AT-DCAEP exhibits excellent performance in multivariate time series data anomaly detection without the need for pre-labeling large-scale datasets, making it an efficient unsupervised anomaly detection method. We extensively tested the performance of AT-DCAEP on six publicly available datasets, and the results show that compared to current state-of-the-art methods, AT-DCAEP demonstrates superior performance, achieving the optimal balance between anomaly detection performance and computational cost.
Aimed at the two problems of underwater imaging, fog effect and color cast, an Improved Segmentation Dark Channel Prior(ISDCP) defogging method is proposed to solve the fog effects caused by physical ...properties of water. Due to mass refraction of light in the process of underwater imaging, fog effects would lead to image blurring. And color cast is closely related to different degree of attenuation while light with different wavelengths is traveling in water. The proposed method here integrates the ISDCP and quantitative histogram stretching techniques into the image enhancement procedure. Firstly, the threshold value is set during the refinement process of the transmission maps to identify the original mismatching, and to conduct the differentiated defogging process further. Secondly, a method of judging the propagating distance of light is adopted to get the attenuation degree of energy during the propagation underwater. Finally, the image histogram is stretched quantitatively in Red-Green-Blue channel respectively according to the degree of attenuation in each color channel. The proposed method ISDCP can reduce the computational complexity and improve the efficiency in terms of defogging effect to meet the real-time requirements. Qualitative and quantitative comparison for several different underwater scenes reveals that the proposed method can significantly improve the visibility compared with previous methods.
This paper presents an approach for improving the stability and adaptability of normal-mode theory, applied to marine environment sound field modeling and calculation. Firstly, based on the custom ...loss function, both Single Output Joint Neural Network (SOJNN) and Multi Output Neural Network (MONN) with physical constraints are constructed to obtain the prediction of wave numbers and eigenfunction. Secondly, we converted the complex wave numbers of normal-mode into a function to find the extreme value, as well as combining the boundary conditions set as the loss function. In this way, the problem space is constrained to the solution space and the dimension of the search space is reduced. Finally, the corresponding sound field is obtained according to the wave numbers and eigenfunction output by neural network. We set the sound velocity at 1500m/s with the density of water is 1g/cm 3 , place the sound source at 32 meters underwater. The upper boundary is absolutely soft and the lower boundary is absolutely hard. The effectiveness of our method is verified by simulation.