Transferable adversarial slow feature extraction network for few-shot quality prediction in coal-to-ethylene glycol process

In the coal-to-ethylene glycol(CTEG)process,precisely estimating quality variables is crucial for process monitoring,optimization,and control.A significant challenge in this regard is relying on offline laboratory analysis to obtain these variables,which often incurs substantial monetary costs and significant time delays.The resulting few-shot learning scenarios present a hurdle to the efficient development of predictive models.To address this issue,our study introduces the transferable adversarial slow feature extraction network(TASF-Net),an innovative approach designed specifically for few-shot quality prediction in the CTEG process.TASF-Net uniquely integrates the slowness principle with a deep Bayesian framework,effectively capturing the nonlinear and inertial characteristics of the CTEG process.Additionally,the model employs a variable attention mechanism to identify quality-related input variables adaptively at each time step.A key strength of TASF-Net lies in its ability to navigate the complex measurement noise,outliers,and system interference typical in CTEG data.Adversarial learning strategy using a min-max game is adopted to improve its robustness and ability to model irregular industrial data accurately and significantly.Furthermore,an incremental refining transfer learning framework is designed to further improve few-shot prediction performance achieved by transferring knowledge from the pretrained model on the source domain to the target domain.The effectiveness and superiority of TASF-Net have been empirically validated using a real-world CTEG dataset.Compared with some state-of-the-art methods,TASF-Net demonstrates exceptional capability in addressing the intricate challenges for few-shot quality prediction in the CTEG process.

Quick Weighing of Passing Vehicles Using the Transfer-Learning-Enhanced Convolutional Neural Network

Transfer learning could reduce the time and resources required by the training of new models and be therefore important for generalized applications of the trainedmachine learning algorithms.In this study,a transfer learningenhanced convolutional neural network(CNN)was proposed to identify the gross weight and the axle weight of moving vehicles on the bridge.The proposed transfer learning-enhanced CNN model was expected to weigh different bridges based on a small amount of training datasets and provide high identification accuracy.First of all,a CNN algorithm for bridge weigh-in-motion(B-WIM)technology was proposed to identify the axle weight and the gross weight of the typical two-axle,three-axle,and five-axle vehicles as they crossed the bridge with different loading routes and speeds.Then,the pre-trained CNN model was transferred by fine-tuning to weigh themoving vehicle on another bridge.Finally,the identification accuracy and the amount of training data required were compared between the two CNN models.Results showed that the pre-trained CNN model using transfer learning for B-WIM technology could be successfully used for the identification of the axle weight and the gross weight for moving vehicles on another bridge while reducing the training data by 63%.Moreover,the recognition accuracy of the pre-trained CNN model using transfer learning was comparable to that of the original model,showing its promising potentials in the actual applications.

IDS-INT:Intrusion detection system using transformer-based transfer learning for imbalanced network traffic

A network intrusion detection system is critical for cyber security against llegitimate attacks.In terms of feature perspectives,network traffic may include a variety of elements such as attack reference,attack type,a subcategory of attack,host information,malicious scripts,etc.In terms of network perspectives,network traffic may contain an imbalanced number of harmful attacks when compared to normal traffic.It is challenging to identify a specific attack due to complex features and data imbalance issues.To address these issues,this paper proposes an Intrusion Detection System using transformer-based transfer learning for Imbalanced Network Traffic(IDS-INT).IDS-INT uses transformer-based transfer learning to learn feature interactions in both network feature representation and imbalanced data.First,detailed information about each type of attack is gathered from network interaction descriptions,which include network nodes,attack type,reference,host information,etc.Second,the transformer-based transfer learning approach is developed to learn detailed feature representation using their semantic anchors.Third,the Synthetic Minority Oversampling Technique(SMOTE)is implemented to balance abnormal traffic and detect minority attacks.Fourth,the Convolution Neural Network(CNN)model is designed to extract deep features from the balanced network traffic.Finally,the hybrid approach of the CNN-Long Short-Term Memory(CNN-LSTM)model is developed to detect different types of attacks from the deep features.Detailed experiments are conducted to test the proposed approach using three standard datasets,i.e.,UNsWNB15,CIC-IDS2017,and NSL-KDD.An explainable AI approach is implemented to interpret the proposed method and develop a trustable model.

Radiative heat transfer analysis of a concave porous fin under the local thermal non-equilibrium condition:application of the clique polynomial method and physics-informed neural networks

The heat transfer through a concave permeable fin is analyzed by the local thermal non-equilibrium(LTNE)model.The governing dimensional temperature equations for the solid and fluid phases of the porous extended surface are modeled,and then are nondimensionalized by suitable dimensionless terms.Further,the obtained nondimensional equations are solved by the clique polynomial method(CPM).The effects of several dimensionless parameters on the fin's thermal profiles are shown by graphical illustrations.Additionally,the current study implements deep neural structures to solve physics-governed coupled equations,and the best-suited hyperparameters are attained by comparison with various network combinations.The results of the CPM and physicsinformed neural network(PINN)exhibit good agreement,signifying that both methods effectively solve the thermal modeling problem.

Neural Network-Based Limiter with Transfer Learning

Recent works have shown that neural networks are promising parameter-free limiters for a variety of numerical schemes(Morgan et al.in A machine learning approach for detect-ing shocks with high-order hydrodynamic methods.et al.in J Comput Phys 367:166-191.,2018;Veiga et al.in European Conference on Computational Mechanics andⅦEuropean Conference on Computational Fluid Dynamics,vol.1,pp.2525-2550.ECCM.,2018).Following this trend,we train a neural network to serve as a shock-indicator function using simulation data from a Runge-Kutta discontinuous Galer-kin(RKDG)method and a modal high-order limiter(Krivodonova in J Comput Phys 226:879-896.,2007).With this methodology,we obtain one-and two-dimensional black-box shock-indicators which are then coupled to a standard limiter.Furthermore,we describe a strategy to transfer the shock-indicator to a residual distribution(RD)scheme without the need for a full training cycle and large data-set,by finding a mapping between the solution feature spaces from an RD scheme to an RKDG scheme,both in one-and two-dimensional problems,and on Cartesian and unstruc-tured meshes.We report on the quality of the numerical solutions when using the neural network shock-indicator coupled to a limiter,comparing its performance to traditional lim-iters,for both RKDG and RD schemes.

Intelligent Fault Diagnosis for Planetary Gearbox Using Transferable Deep Q Network Under Variable Conditions with Small Training Data

Effective fault diagnosis of planetary gearboxes is critical for ensuring the safety and dependability of mechanical drive systems.Nevertheless,variable conditions and inadequate fault data bring huge challenges to its practical fault diagnosis.Taking this into account,this study presents a new intelligent fault diagnosis(IFD)approach for planetary gearbox using a transferable deep Q network(TDQN)that merges deep reinforcement learning(DRL)and transfer learning(TL).First,a DRL environment simulation is designed by a predefined classification Markov decision process.Then,leveraging varied-size convolutions and residual learning,a multiscale residual convolutional neural network agent for TDQN is created to automatically learn meaningful features directly from vibration signals while avoiding model degradation.Next,a large source dataset is obtained from complex conditions,and this agent learns an IFD policy via autonomous interaction with the data environment.Finally,a parameter-based TL strategy is adopted to retrain the model on target datasets with variable conditions and small training data,which is conducted by fine-tuning the model parameters gained from the source task to accomplish target tasks.The results show that this TDQN outperforms not only state-of-the-art methods in a source task with an accuracy of 98.53%but also in two target tasks with 99.63%and 98.37%,respectively.

APST-Flow: A Reversible Network-Based Artistic Painting Style Transfer Method

In recent years,deep generative models have been successfully applied to perform artistic painting style transfer(APST).The difficulties might lie in the loss of reconstructing spatial details and the inefficiency of model convergence caused by the irreversible en-decoder methodology of the existing models.Aiming to this,this paper proposes a Flow-based architecture with both the en-decoder sharing a reversible network configuration.The proposed APST-Flow can efficiently reduce model uncertainty via a compact analysis-synthesis methodology,thereby the generalization performance and the convergence stability are improved.For the generator,a Flow-based network using Wavelet additive coupling(WAC)layers is implemented to extract multi-scale content features.Also,a style checker is used to enhance the global style consistency by minimizing the error between the reconstructed and the input images.To enhance the generated salient details,a loss of adaptive stroke edge is applied in both the global and local model training.The experimental results show that the proposed method improves PSNR by 5%,SSIM by 6.2%,and decreases Style Error by 29.4%over the existing models on the ChipPhi set.The competitive results verify that APST-Flow achieves high-quality generation with less content deviation and enhanced generalization,thereby can be further applied to more APST scenes.

Transfer Learning-Based Semi-Supervised Generative Adversarial Network for Malaria Classification

Malaria is a lethal disease responsible for thousands of deaths worldwide every year.Manual methods of malaria diagnosis are timeconsuming that require a great deal of human expertise and efforts.Computerbased automated diagnosis of diseases is progressively becoming popular.Although deep learning models show high performance in the medical field,it demands a large volume of data for training which is hard to acquire for medical problems.Similarly,labeling of medical images can be done with the help of medical experts only.Several recent studies have utilized deep learning models to develop efficient malaria diagnostic system,which showed promising results.However,the most common problem with these models is that they need a large amount of data for training.This paper presents a computer-aided malaria diagnosis system that combines a semi-supervised generative adversarial network and transfer learning.The proposed model is trained in a semi-supervised manner and requires less training data than conventional deep learning models.Performance of the proposed model is evaluated on a publicly available dataset of blood smear images(with malariainfected and normal class)and achieved a classification accuracy of 96.6%.

NOMA Empowered Energy Efficient Data Collection and Wireless Power Transfer in Space-Air-Ground Integrated Networks

As the sixth generation network(6G)emerges,the Internet of remote things(IoRT)has become a critical issue.However,conventional terrestrial networks cannot meet the delay-sensitive data collection needs of IoRT networks,and the Space-Air-Ground integrated network(SAGIN)holds promise.We propose a novel setup that integrates non-orthogonal multiple access(NOMA)and wireless power transfer(WPT)to collect latency-sensitive data from IoRT networks.To extend the lifetime of devices,we aim to minimize the maximum energy consumption among all IoRT devices.Due to the coupling between variables,the resulting problem is non-convex.We first decouple the variables and split the original problem into four subproblems.Then,we propose an iterative algorithm to solve the corresponding subproblems based on successive convex approximation(SCA)techniques and slack variables.Finally,simulation results show that the NOMA strategy has a tremendous advantage over the OMA scheme in terms of network lifetime and energy efficiency,providing valuable insights.

Design space exploration of neural network accelerator based on transfer learning

With the increasing demand of computational power in artificial intelligence(AI)algorithms,dedicated accelerators have become a necessity.However,the complexity of hardware architectures,vast design search space,and complex tasks of accelerators have posed significant challenges.Tra-ditional search methods can become prohibitively slow if the search space continues to be expanded.A design space exploration(DSE)method is proposed based on transfer learning,which reduces the time for repeated training and uses multi-task models for different tasks on the same processor.The proposed method accurately predicts the latency and energy consumption associated with neural net-work accelerator design parameters,enabling faster identification of optimal outcomes compared with traditional methods.And compared with other DSE methods by using multilayer perceptron(MLP),the required training time is shorter.Comparative experiments with other methods demonstrate that the proposed method improves the efficiency of DSE without compromising the accuracy of the re-sults.

Resting-state brain network remodeling after different nerve reconstruction surgeries:a functional magnetic resonance imaging study in brachial plexus injury rats

Distinct brain remodeling has been found after different nerve reconstruction strategies,including motor representation of the affected limb.However,differences among reconstruction strategies at the brain network level have not been elucidated.This study aimed to explore intranetwork changes related to altered peripheral neural pathways after different nerve reconstruction surgeries,including nerve repair,endto-end nerve transfer,and end-to-side nerve transfer.Sprague–Dawley rats underwent complete left brachial plexus transection and were divided into four equal groups of eight:no nerve repair,grafted nerve repair,phrenic nerve end-to-end transfer,and end-to-side transfer with a graft sutured to the anterior upper trunk.Resting-state brain functional magnetic resonance imaging was obtained 7 months after surgery.The independent component analysis algorithm was utilized to identify group-level network components of interest and extract resting-state functional connectivity values of each voxel within the component.Alterations in intra-network resting-state functional connectivity were compared among the groups.Target muscle reinnervation was assessed by behavioral observation(elbow flexion)and electromyography.The results showed that alterations in the sensorimotor and interoception networks were mostly related to changes in the peripheral neural pathway.Nerve repair was related to enhanced connectivity within the sensorimotor network,while end-to-side nerve transfer might be more beneficial for restoring control over the affected limb by the original motor representation.The thalamic-cortical pathway was enhanced within the interoception network after nerve repair and end-to-end nerve transfer.Brain areas related to cognition and emotion were enhanced after end-to-side nerve transfer.Our study revealed important brain networks related to different nerve reconstructions.These networks may be potential targets for enhancing motor recovery.

MDCN:Modified Dense Convolution Network Based Disease Classification in Mango Leaves

The most widely farmed fruit in the world is mango.Both the production and quality of the mangoes are hampered by many diseases.These diseases need to be effectively controlled and mitigated.Therefore,a quick and accurate diagnosis of the disorders is essential.Deep convolutional neural networks,renowned for their independence in feature extraction,have established their value in numerous detection and classification tasks.However,it requires large training datasets and several parameters that need careful adjustment.The proposed Modified Dense Convolutional Network(MDCN)provides a successful classification scheme for plant diseases affecting mango leaves.This model employs the strength of pre-trained networks and modifies them for the particular context of mango leaf diseases by incorporating transfer learning techniques.The data loader also builds mini-batches for training the models to reduce training time.Finally,optimization approaches help increase the overall model’s efficiency and lower computing costs.MDCN employed on the MangoLeafBD Dataset consists of a total of 4,000 images.Following the experimental results,the proposed system is compared with existing techniques and it is clear that the proposed algorithm surpasses the existing algorithms by achieving high performance and overall throughput.

Evaluation of the famous classic formula Sanhua decoction based on network pharmacology and multi-component quantitative analysis

Background:Sanhua decoction has significant effects in the treatment of stroke.The study of the Sanhua decoction material benchmark was carried out to analyze the value transfer relationship between the Chinese herbal pieces and the substance benchmark.Methods:Network pharmacology was employed to investigate the potential active components and molecular mechanisms of Sanhua decoction in the treatment of stroke.15 batches of Sanhua decoction lyophilized powder were prepared using traditional formulas and subjected to high-performance liquid chromatography analysis to generate fingerprints of the Sanhua decoction substance benchmarks.Then,a multi-component quantitative analysis method was established,allowing for the simultaneous determination of ten components,to study the transfer of quantity values between pieces and substance benchmarks.Results:60 active ingredients were screened from Sanhua decoction by network pharmacology,of which gallic acid,magnolol honokiol,physcion,and aloe-emodin may have a greater effect than other active components.63 key targets and 134 pathways were predicted as the potential mechanism of Sanhua decoction in treating stroke.The fingerprint similarity of the Sanhua decoction substance benchmarks was found to be good among the 15 batches,confirming the 19 common peaks.The content of the 10 components was basically consistent.The components’transfer rates were within 30%of their respective means.Conclusions:This study provided a comprehensive and reliable strategy for the quality evaluation of Sanhua decoction substance benchmarks and held significant importance in improving its application value.

Deep Learning-Based Mask Identification System Using ResNet Transfer Learning Architecture

Recently,the coronavirus disease 2019 has shown excellent attention in the global community regarding health and the economy.World Health Organization(WHO)and many others advised controlling Corona Virus Disease in 2019.The limited treatment resources,medical resources,and unawareness of immunity is an essential horizon to unfold.Among all resources,wearing a mask is the primary non-pharmaceutical intervention to stop the spreading of the virus caused by Severe Acute Respiratory Syndrome Coronavirus 2(SARS-CoV-2)droplets.All countries made masks mandatory to prevent infection.For such enforcement,automatic and effective face detection systems are crucial.This study presents a face mask identification approach for static photos and real-time movies that distinguishes between images with and without masks.To contribute to society,we worked on mask detection of an individual to adhere to the rule and provide awareness to the public or organization.The paper aims to get detection accuracy using transfer learning from Residual Neural Network 50(ResNet-50)architecture and works on detection localization.The experiment is tested with other popular pre-trained models such as Deep Convolutional Neural Networks(AlexNet),Residual Neural Networks(ResNet),and Visual Geometry Group Networks(VGG-Net)advanced architecture.The proposed system generates an accuracy of 98.4%when modeled using Residual Neural Network 50(ResNet-50).Also,the precision and recall values are proved as better when compared to the existing models.This outstanding work also can be used in video surveillance applications.

Slope displacement prediction based on multisource domain transfer learning for insufficient sample data

Accurate displacement prediction is critical for the early warning of landslides.The complexity of the coupling relationship between multiple influencing factors and displacement makes the accurate prediction of displacement difficult.Moreover,in engineering practice,insufficient monitoring data limit the performance of prediction models.To alleviate this problem,a displacement prediction method based on multisource domain transfer learning,which helps accurately predict data in the target domain through the knowledge of one or more source domains,is proposed.First,an optimized variational mode decomposition model based on the minimum sample entropy is used to decompose the cumulative displacement into the trend,periodic,and stochastic components.The trend component is predicted by an autoregressive model,and the periodic component is predicted by the long short-term memory.For the stochastic component,because it is affected by uncertainties,it is predicted by a combination of a Wasserstein generative adversarial network and multisource domain transfer learning for improved prediction accuracy.Considering a real mine slope as a case study,the proposed prediction method was validated.Therefore,this study provides new insights that can be applied to scenarios lacking sample data.

Automatic detection of breast lesions in automated 3D breast ultrasound with cross-organ transfer learning

Background Deep convolutional neural networks have garnered considerable attention in numerous machine learning applications,particularly in visual recognition tasks such as image and video analyses.There is a growing interest in applying this technology to diverse applications in medical image analysis.Automated three dimensional Breast Ultrasound is a vital tool for detecting breast cancer,and computer-assisted diagnosis software,developed based on deep learning,can effectively assist radiologists in diagnosis.However,the network model is prone to overfitting during training,owing to challenges such as insufficient training data.This study attempts to solve the problem caused by small datasets and improve model detection performance.Methods We propose a breast cancer detection framework based on deep learning(a transfer learning method based on cross-organ cancer detection)and a contrastive learning method based on breast imaging reporting and data systems(BI-RADS).Results When using cross organ transfer learning and BIRADS based contrastive learning,the average sensitivity of the model increased by a maximum of 16.05%.Conclusion Our experiments have demonstrated that the parameters and experiences of cross-organ cancer detection can be mutually referenced,and contrastive learning method based on BI-RADS can improve the detection performance of the model.

Transfer Learning Approach to Classify the X-Ray Image that Corresponds to Corona Disease Using ResNet50 Pre-Trained by ChexNet

The COVID-19 pandemic has had a widespread negative impact globally. It shares symptoms with other respiratory illnesses such as pneumonia and influenza, making rapid and accurate diagnosis essential to treat individuals and halt further transmission. X-ray imaging of the lungs is one of the most reliable diagnostic tools. Utilizing deep learning, we can train models to recognize the signs of infection, thus aiding in the identification of COVID-19 cases. For our project, we developed a deep learning model utilizing the ResNet50 architecture, pre-trained with ImageNet and CheXNet datasets. We tackled the challenge of an imbalanced dataset, the CoronaHack Chest X-Ray dataset provided by Kaggle, through both binary and multi-class classification approaches. Additionally, we evaluated the performance impact of using Focal loss versus Cross-entropy loss in our model.

Enhancing Pneumonia Detection in Pediatric Chest X-Rays Using CGAN-Augmented Datasets and Lightweight Deep Transfer Learning Models

Pneumonia ranks as a leading cause of mortality, particularly in children aged five and under. Detecting this disease typically requires radiologists to examine chest X-rays and report their findings to physicians, a task susceptible to human error. The application of Deep Transfer Learning (DTL) for the identification of pneumonia through chest X-rays is hindered by a shortage of available images, which has led to less than optimal DTL performance and issues with overfitting. Overfitting is characterized by a model’s learning that is too closely fitted to the training data, reducing its effectiveness on unseen data. The problem of overfitting is especially prevalent in medical image processing due to the high costs and extensive time required for image annotation, as well as the challenge of collecting substantial datasets that also respect patient privacy concerning infectious diseases such as pneumonia. To mitigate these challenges, this paper introduces the use of conditional generative adversarial networks (CGAN) to enrich the pneumonia dataset with 2690 synthesized X-ray images of the minority class, aiming to even out the dataset distribution for improved diagnostic performance. Subsequently, we applied four modified lightweight deep transfer learning models such as Xception, MobileNetV2, MobileNet, and EfficientNetB0. These models have been fine-tuned and evaluated, demonstrating remarkable detection accuracies of 99.26%, 98.23%, 97.06%, and 94.55%, respectively, across fifty epochs. The experimental results validate that the models we have proposed achieve high detection accuracy rates, with the best model reaching up to 99.26% effectiveness, outperforming other models in the diagnosis of pneumonia from X-ray images.

Research on Rotating Machinery Fault Diagnosis Based on Improved Multi-target Domain Adversarial Network

Aiming at the problems of low efficiency,poor anti-noise and robustness of transfer learning model in intelligent fault diagnosis of rotating machinery,a new method of intelligent fault diagnosis of rotating machinery based on single source and multi-target domain adversarial network model(WDMACN)and Gram Angle Product field(GAPF)was proposed.Firstly,the original one-dimensional vibration signal is preprocessed using GAPF to generate the image data including all time series.Secondly,the residual network is used to extract data features,and the features of the target domain without labels are pseudo-labeled,and the transferable features among the feature extractors are shared through the depth parameter,and the feature extractors of the multi-target domain are updated anatomically to generate the features that the discriminator cannot distinguish.The modelt through adversarial domain adaptation,thus achieving fault classification.Finally,a large number of validations were carried out on the bearing data set of Case Western Reserve University(CWRU)and the gear data.The results show that the proposed method can greatly improve the diagnostic efficiency of the model,and has good noise resistance and generalization.

A priority-based dynamic load transfer algorithm for cellular/WLAN integrated networks

For the integration network of a cellular network and a wireless local area network （WLAN）, a priority-based dynamic load transfer （PDLT） algorithm is proposed. The dynamic vertical handoffs by call admission control are jointly determined by the network conditions and the traffic characteristics in combination with the location-condition of mobile terminals. When there is no bandwidth resource available in the cellular network or WLAN, the proposed PDLT algorithm allows an incoming voice call or data call within the overlapping area of the cellular network and the WLAN to be directed to the spare network; meanwhile, by dynamically computing the occupancy of the bandwidth resource, the proposed PDLT algorithm also allows an ongoing voice call or data communication to be transferred to the network with a sufficient bandwidth resource according to the given threshold to balance the number of voice/data calls in the two networks. The analysis results of a two-dimensional Markov model and the simulation results show that the PDLT algorithm can effectively enhance the whole integrated network＇ s traffic, reduce the blocking probability of new calls and increase the data throughput, and thus decrease the response time for various services.