Search

With the development of emerging information technology, the traditional management methods of marine fishes are slowly replaced by new methods due to high cost, time-consumption and inaccurate management. The update of marine fishes management technology is also a great help for the creation of smart cities. However, some new methods have been studied that are too specific, which are not applicable for the other marine fishes, and the accuracy of identification is generally low. Therefore, this paper proposes an ecological Internet of Things (IoT) framework, in which a lightweight Deep Neural Networks model is implemented as a image recognition model for marine fishes, which is recorded as Fish-CNN. In this study, multi-training and evaluation of Fish-CNN is accomplished, and the accuracy of the final classification can be fixed to 89.89%–99.83%. Moreover, the final evaluation compared with Rem-CNN, Linear Regression and Multilayer Perceptron also verify the stability and advantage of our method. [ABSTRACT FROM AUTHOR]

Desynchronization attacks proved to be the greatest challenge to audio watermarking systems as they introduce misalignment between the signal carrier and the watermark. This paper proposes a DNN-based speech watermarking system with two adversarial networks jointly trained on a set of desynchronization attacks to embed a randomly generated watermark. The detector neural network is expanded with spatial pyramid pooling layers to be able to handle signals affected by these attacks. A detailed training procedure of the aforementioned DNN system with gradual attack introduction is proposed in order to achieve robustness. Experiments performed on a speech dataset show that the system achieves satisfactory results according to all the benchmarks it was tested against. The system preserves signal quality after watermark embedding. Most importantly, the system achieved resistance to all considered desynchronization attacks. The majority of the attacks cause less than 1. 7 0 % of incorrectly detected watermarked bits on average, which outperforms comparative techniques in this regard. [ABSTRACT FROM AUTHOR]

Retail demand forecasting is a highly intricate and multilevel problem as it may include numerous products, under multiple categories, and potentially sold in many different stores. Forecasts may also process products in isolation or collectively so as to detect mutual correlations. In technical terms, retail demand forecasting may be broken down to a multivariate time series forecasting problem where the time axis is prevalent and introduces added complexity which necessitates proper usage of advanced methodologies with machine/deep learning backgrounds. Popular boosting regressors, such as XGBoost and LightGBM, are excellent machine learning candidates with extensive bibliographic backgrounds covering multiple scientific fields and applications including generic multivariate time series forecasts. The Temporal Convolutional Network and the Temporal Fusion Transformer are recently proposed deep neural network architectures that found success in specific multivariate scenarios but are yet to be tested in a variety of related fields such as retail demand forecasting. Therefore, within this paper, these novel in-the-field models are compared to the aforementioned boosting approaches alongside popular statistical univariate methods, namely Exponential Smoothing and SARIMA. To properly compare the selected models and attempt to generalize their usability, two separate datasets are analyzed and forecasted; an item sales dataset with 500 time series and a category sales dataset with 540 time series. The findings suggest that the deep learning solutions are the better predictors, with the Transformer model surpassing the Boosting solutions by up to 16.8% sMAPE decrease and the statistical approaches by up to 28.6% decrease, further substantiating the notion that deep learning techniques are exceptionally promising in handling large-scale, non-linear and outlier data whilst suggesting that retail demand forecasting does benefit from multivariate approaches and the application of advanced machine learning methods. [ABSTRACT FROM AUTHOR]

In the resource-constrained embedded systems, the designing of efficient deep neural networks is a challenging process, due to diversity in the artificial intelligence applications. The quantization in deep neural networks superiorly diminishes the storage and computational time by reducing the bit-width of networks encoding. In order to highlight the problem of accuracy loss, the quantization levels are automatically discovered using Policy Feedback-based Reinforcement Learning Method (PF-RELEQ). In this paper, the Proximal Policy Optimization with Policy Feedback (PPO-PF) technique is proposed to determine the best design decisions by choosing the optimum hyper-parameters. In order to enhance the sensitivity of the value function to the change of policy and to improve the accuracy of value estimation at the early learning stage, a policy update method is devised based on the clipped discount factor. In addition, specifically the loss functions of policy satisfy the unbiased estimation of the trust region. The proposed PF-RELEQ effectively balances quality and speed compared to other deep learning methods like ResNet-1202, ResNet-32, ResNet-110, GoogLeNet and AlexNet. The experimental analysis showed that PF-RELEQ achieved 20% computational work-load reduction compared to the existing deep learning methods on ImageNet, CIFAR-10, CIFAR-100 and tomato leaf disease datasets and achieved approximately 2% of improvisation in the validation accuracy. Additionally, the PF-RELEQ needs only 0.55 Graphics Processing Unit on an NVIDIA GTX-1080Ti to develop DNNs that delivers better accuracy improvement with fewer cycle counts for image classification. [ABSTRACT FROM AUTHOR]

Identifying a core set of features is one of the most important steps in the development of an automated seizure detector. In most of the published studies describing features and seizure classifiers, the features were hand-engineered, which may not be optimal. The main goal of the present paper is using deep convolutional neural networks (CNNs) and random forest to automatically optimize feature selection and classification. The input of the proposed classifier is raw multi-channel EEG and the output is the class label: seizure/nonseizure. By training this network, the required features are optimized, while fitting a nonlinear classifier on the features. After training the network with EEG recordings of 26 neonates, five end layers performing the classification were replaced with a random forest classifier in order to improve the performance. This resulted in a false alarm rate of 0.9 per hour and seizure detection rate of 77% using a test set of EEG recordings of 22 neonates that also included dubious seizures. The newly proposed CNN classifier outperformed three data-driven feature-based approaches and performed similar to a previously developed heuristic method. [ABSTRACT FROM AUTHOR]

Domain adaptation aims to generalize the classification model from a source domain to a different but related target domain. Recent studies have revealed the benefit of deep convolutional features trained on a large dataset (e.g. ImageNet) in alleviating domain discrepancy. However, literatures show that the transferability of features decreases as (i) the difference between the source and target domains increases, or (ii) the layers are toward the top layers. Therefore, even with deep features, domain adaptation remains necessary. In this paper, we propose a novel unsupervised domain adaptation (UDA) model for deep neural networks, which is learned with the labeled source samples and the unlabeled target ones simultaneously. For target samples without labels, pseudo labels are assigned to them according to their maximum classification scores during training of the UDA model. However, due to the domain discrepancy, label noise generally is inevitable, which degrades the performance of the domain adaptation model. Thus, to effectively utilize the target samples, three specific robust deep softmax regression (RDSR) functions are performed for them with high, medium and low classification confidence respectively. Extensive experiments show that our method yields the state-of-the-art results, demonstrating the effectiveness of the robust deep softmax regression classifier in UDA. [ABSTRACT FROM AUTHOR]

Cardiomegaly is a radiographic abnormality, and it has significant prognosis importance in the population. Chest X-ray images can identify it. Early detection of cardiomegaly reduces the risk of congestive heart failure and systolic dysfunction. Due to the lack of radiologists, there is a demand for the artificial intelligence tool for the early detection of cardiomegaly. The cardiomegaly X-ray dataset is extracted from the cheXpert database. Totally, 46195 X-ray records with a different view such as AP view, PA views, and lateral views are used to train and validate the proposed model. The artificial intelligence app named CardioXpert is constructed based on deep neural network. The transfer learning approach is adopted to increase the prediction metrics, and an optimized training method called adaptive movement estimation is used. Three different transfer learning-based deep neural networks named APNET, PANET, and LateralNET are constructed for each view of X-ray images. Finally, certainty-based fusion is performed to enrich the prediction accuracy, and it is named CardioXpert. As the proposed method is based on the largest cardiomegaly dataset, hold-out validation is performed to verify the prediction accuracy of the proposed model. An unseen dataset validates the model. These deep neural networks, APNET, PANET, and LateralNET, are individually validated, and then the fused network CardioXpert is validated. The proposed model CardioXpert provides an accuracy of 93.6%, which is the highest at this time for this dataset. It also yields the highest sensitivity of 94.7% and a precision of 97.7%. These prediction metrics prove that the proposed model outperforms all the state-of-the-art deep transfer learning methods for diagnosing cardiomegaly thoracic disorder. The proposed deep learning neural network model is deployed as the web app. The cardiologist can use this prognostic app to predict cardiomegaly disease faster and more robust in the early state by using low-cost and chest X-ray images. [ABSTRACT FROM AUTHOR]

Using second-order optimization methods for training deep neural networks (DNNs) has attracted many researchers. A recently proposed method, Eigenvalue-corrected Kronecker Factorization (EKFAC), proposed an interpretation by viewing natural gradient update as a diagonal method and corrects the inaccurate re-scaling factor in the KFAC eigenbasis. What's more, a new method to approximate the natural gradient called Trace-restricted Kronecker-factored Approximate Curvature (TKFAC) is also proposed, in which the Fisher information matrix (FIM) is approximated as a constant multiplied by the Kronecker product of two matrices and the traces can be kept equal before and after the approximation. In this work, we combine the ideas of these two methods and propose Trace-restricted Eigenvalue-corrected Kronecker Factorization (TEKFAC). The proposed method not only corrects the inexact re-scaling factor under the Kronecker-factored eigenbasis, but also considers the new approximation method and the effective damping technique adopted by TKFAC. We also discuss the differences and relationships among the related Kronecker-factored approximations. Empirically, our method outperforms SGD with momentum, Adam, EKFAC and TKFAC on several DNNs. [ABSTRACT FROM AUTHOR]

This study aims to adapt the Expectation Disconfirmation Theory and Technology Adoption Model to unveil provocative roles in patients' satisfaction cognitions and subsequent continuity behaviors pertaining to telemedicine services in rural Bangladesh. A quantitative research model is developed and validated using a two-staged deep neural network and partial least squares structural equation modeling approach. The findings of this study provide evidence that five salient determinants; expectations, disconfirmation, performance, usefulness, and ease of use dominantly contribute to predicting patients' satisfaction concerning continuity with telemedicine. This contributes to health informatics and behavioral literature by clarifying the complex interplay between patients' satisfaction and determinants of continuity behavior in telemedicine's domain. The findings provide novel insights into predictions of complex patients' attitudes toward telemedicine continuity, and dynamic changes in adoption trends thereby assisting health professionals, global health experts, policymakers, and IS community in making higher quality informed decisions for people-centered future models of care. [ABSTRACT FROM AUTHOR]

Improving the reliability of nighttime pedestrian detection is a crucial challenge towards the design of robust autonomous systems. Not surprisingly, most pedestrian fatalities occur in low-illumination settings, thus emphasizing the need for new algorithmic advances. This work presents a novel pedestrian detection approach that makes a number of crucial modifications to the state-of-the-art YOLOV5-PANet architecture, in order to improve the reliability of features extracted from nighttime images. More specifically, the proposed architecture systematically incorporates powerful shuffle attention mechanisms and a transformer module to improve the feature learning pipeline. Instead of advocating the use of other sensing modalities that are better suited for nighttime detection, our approach relies only on conventional RGB cameras and is hence broadly applicable. Our empirical studies with nighttime pedestrian detection benchmarks show that with only minimal increase in model complexity, our approach provides significant improvements in detection efficacy over existing solutions. Finally, we explore the impact of post-hoc network pruning on the speed-accuracy trade-off of our approach and demonstrate that it is well suited for reduced memory/compute requirements. [ABSTRACT FROM AUTHOR]