检索结果-南通市图书馆

Identifying malicious traffic under concept drift based on intraclass consistency enhanced variational autoencoder

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第8期67卷 238-252页

作者： Xiang LUO Chang liu Gaopeng GOU Gang XIONG zhen LI Binxing FANG Institute of Information Engineering Chinese Academy of Sciences School of Cyber Security University of Chinese Academy of Sciences School of Computer Science and Technology Harbin Institute of Technology (Shenzhen)

Accurate identification of malicious traffic is crucial for implementing effective defense countermeasures and has led to extensive research efforts. However, the continuously evolving techniques employed by adversaries have introduced the issues of concept drift, which significantly affects the performance of existing methods. To tackle this challenge, some researchers have focused on improving the separability of malicious traffic representation and designing drift detectors to reduce the number of false ***, these methods often overlook the importance of enhancing the generalization and intraclass consistency in the representation. Additionally, the detectors are not sufficiently sensitive to the variations among different malicious traffic classes, which results in poor performance and limited robustness. In this paper, we propose intraclass consistency enhanced variational autoencoder with Class-Perception detector(ICE-CP) to identify malicious traffic under concept drift. It comprises two key modules during training:intraclass consistency enhanced(ICE) representation learning and Class-Perception(CP) detector construction. In the first module, we employ a variational autoencoder(VAE) in conjunction with Kullback-Leibler(KL)-divergence and cross-entropy loss to model the distribution of each input malicious traffic flow. This approach simultaneously enhances the generalization, interclass consistency, and intraclass differences in the learned representation. Consequently, we obtain a compact representation and a trained classifier for nondrifting malicious traffic. In the second module, we design the CP detector, which generates a centroid and threshold for each malicious traffic class separately based on the learned representation, depicting the boundaries between drifting and non-drifting malicious traffic. During testing, we utilize the trained classifier to predict malicious traffic classes for the testing samples. Then, we use the CP det

关键词： concept drift cyberspace security intrusion detection malicious traffic identification variational autoencoder

同方期刊数据库

SDCC: software-defined collective communication for distributed training

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第9期67卷 105-125页

作者： Xin JIN zhen ZHANG Yunshan JIA Yun MA Xuanzhe liu School of Computer Science Peking University Department of Computer Science and Technology Johns Hopkins University

Communication is crucial to the performance of distributed training. Today's solutions tightly couple the control and data planes and lack flexibility, generality, and performance. In this study, we present SDCC, a software-defined collective communication framework for distributed training. SDCC is based on the principle of modern systems design to effectively decouple the control plane from the data *** abstracts the operations for collective communication in distributed training with dataflow operations and unifies computing and communication with a single dataflow graph. The abstraction, together with the unification, is powerful: it enables users to easily express new and existing collective communication algorithms and optimizations, simplifies the integration with different computing engines(e.g., Py Torch and Tensor Flow) and network transports(e.g., Linux TCP and kernel bypass), and allows the system to improve performance by exploiting parallelism exposed by the dataflow graph. We further demonstrate the benefits of SDCC in four use cases.

关键词： machine learning systems distributed training deep learning collective communication software-defined networking

同方期刊数据库

Cell reprogramming therapy for Parkinson’s disease

在线全文

学校读者我要写书评

暂无评论

Neural Regeneration Research 2024年第11期19卷 2444-2455页

作者： Wenjing dong Shuyi liu Shangang Li zhengbo Wang State Key Laboratory of Primate Biomedical Research Institute of Primate Translational MedicineKunming University of Science and TechnologyKunmingYunnan ProvinceChina Yunnan Key Laboratory of Primate Biomedical Research KunmingYunnan ProvinceChina

Parkinson’s disease is typically characterized by the progressive loss of dopaminergic neurons in the substantia nigra pars *** studies have been performed based on the supplementation of lost dopaminergic neurons to treat Parkinson’s *** initial strategy for cell replacement therapy used human fetal ventral midbrain and human embryonic stem cells to treat Parkinson’s disease,which could substantially alleviate the symptoms of Parkinson’s disease in clinical ***,ethical issues and tumor formation were limitations of its clinical *** pluripotent stem cells can be acquired without sacrificing human embryos,which eliminates the huge ethical barriers of human stem cell *** widely considered neuronal regeneration strategy is to directly reprogram fibroblasts and astrocytes into neurons,without the need for intermediate proliferation states,thus avoiding issues of immune rejection and tumor *** induced pluripotent stem cells and direct reprogramming of lineage cells have shown promising results in the treatment of Parkinson’s ***,there are also ethical concerns and the risk of tumor formation that need to be *** review highlights the current application status of cell reprogramming in the treatment of Parkinson’s disease,focusing on the use of induced pluripotent stem cells in cell replacement therapy,including preclinical animal models and progress in clinical *** review also discusses the advancements in direct reprogramming of lineage cells in the treatment of Parkinson’s disease,as well as the controversy surrounding in vivo *** findings suggest that cell reprogramming may hold great promise as a potential strategy for treating Parkinson’s disease.

关键词： animal models astrocytes autologous cell reprogramming cell therapy direct lineage reprogramming dopaminergic neurons induced pluripotent stem cells non-human primates Parkinson’s disease

Microstructure and properties of LZQT600-3 HCCDIBs for plunger pump cylinder

在线全文

学校读者我要写书评

暂无评论

China Foundry 2024年第2期21卷 197-204页

作者： Chun-jie Xu Yuan-ying Jin dong Ma zhen Zhao Jia-wei Qi Shang Sui Xiang-quan Wu Can Guo Zhong-ming Zhang Yong-hui liu Dan Shechtman School of Materials Science and Engineering Xi'an University of TechnologyXi'an 710048China Xi'an Shechtman Nobel Prize New Materials Institute Xi'an 710048China Tongchuan Huazhu Precision Technology Co. Ltd.Tongchuan 727100ShaanxiChina Faculty of Materials Science and Engineering Technion-Israel Insititue of TechnologyHaifa 3200003Israel

It is important to improve the comprehensive performance of the ductile iron bars(DIBs)for the cylinder block of the extra high pressure hydraulic plunger pump and accelerate the industrial *** this work,the LZQT600-3 DIBs with the diameter of 145 mm were prepared by the horizontal continuous casting(HCC)process,that is,LZQT600-3 *** microstructure and room temperature tensile properties of different sections[left-edge(surface layer),left-1/2R(left half of the radius),and the center of the HCCDIBs]were *** results show that the spheroidization of LZQT600-3 HCCDIBs matrix from the left-edge,left-1/2R to the center is at nodulizing grade II and *** the cooling rate gradually decreases from surface to the center of the HCCIBs,the number of spheroidized graphite is gradually reduced,the size is gradually increased,the shape factor is decreased,and the pearlite content and lamellate spacing are *** the horizontal direction of the section,the hardness of the material is distributed symmetrically around the center of the *** the vertical direction,the hardness distribution in the center of the HCCDIBs is asymmetrical due to the gravity during the solidification ***,the microstructure in the lower part of the section solidifies relatively *** left-edge has the best tensile mechanical properties,and the ultimate tensile strength,yield tensile strength and elongation are 597.3 MPa,418.5 MPa and 9.6%,*** tensile fracture belongs to the ductile-brittle hybrid *** comprehensive performances of LZQT600-3 HCCDIBs meet the actual application requirements of ultra-high pressure hydraulic plunger pump cylinder.

关键词： horizontal continuous casting(HCC) nodular cast iron ductile iron bars microstructure mechanical properties

Credit assignment for trained neural networks based on Koopman operator theory

在线全文

学校读者我要写书评

暂无评论

Frontiers of Computer Science 2024年第1期18卷 251-253页

作者： zhen LIANG Changyuan ZHAO Wanwei liu Bai XUE Wenjing YANG zhengbin PANG Institute for Quantum Information&State Key Laboratory of High Performance Computing National University of Defense TechnologyChangsha 410000China Institute of Software Chinese Academy of SciencesBeijing 100190China College of Computer Science and Technology National University of Defense TechnologyChangsha 410000China

1 Introduction Artificial neural networks(ANNs,also NNs)have recently emerged as leading candidate models for deep learning,popularly used in various areas[1–3].Behind the enormous success,ANNs are generally with complicated structures,there being an intricate data flow through multiple linear or nonlinear components between the input layer and the output ***,it is pressing to evaluate how much a specific component contributes to the final output,termed the Credit Assignment Problem(CAP)[4]in this paper.

关键词： layer. assignment operator

Analysis and Modeling of Time Output Characteristics for Distributed Photovoltaic and Energy Storage

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Energy Engineering 2024年第4期121卷 933-949页

作者： Kaicheng liu Chen Liang Xiaoyang dong Liping liu China Electric Power Research Institute Beijing100192China Electric Power Research Institute State Grid Gansu Electric Power CompanyLanzhou730000China State Key Laboratory of Power Grid Safety Beijing100084China

Due to the unpredictable output characteristics of distributed photovoltaics,their integration into the grid can lead to voltage fluctuations within the regional power ***,the development of spatial-temporal coordination and optimization control methods for distributed photovoltaics and energy storage systems is of utmost importance in various *** paper approaches the issue from the perspective of spatiotemporal forecasting of distributed photovoltaic(PV)generation and proposes a Temporal Convolutional-Long Short-Term Memory prediction model that combines Temporal Convolutional Networks(TCN)and Long Short-Term Memory(LSTM).To begin with,an analysis of the spatiotemporal distribution patterns of PV generation is conducted,and outlier data is handled using the 3σ ***,a novel approach that combines temporal convolution and LSTM networks is introduced,with TCN extracting spatial features and LSTM capturing temporal ***,a real spatiotemporal dataset from Gansu,China,is established to compare the performance of the proposed network against other *** results demonstrate that the model presented in this paper exhibits the highest predictive accuracy,with a single-step Mean Absolute Error(MAE)of 1.782 and an average Root Mean Square Error(RMSE)of 3.72 for multi-step predictions.

关键词： Photovoltaic power generation spatio-temporal prediction temporal convolutional network long short-term memory network

Weighted Composition Operators That Preserve Frames On Weighted Hardy Spaces

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Acta Mathematica Sinica,English Series 2024年第7期40卷 1645-1654页

作者： Ya Li dong Rui liu Guo Xiang Lu School of Mathematical Sciences and LPMC Nankai UniversityTianjin300071P.R.China School of Statistics and Mathematics Zhongnan University of Economics and LawWuhan430073P.R.China

In this paper, we characterize weighted composition operators that preserve frames on the weighted Hardy spaces in the unit disk. In particular, we obtain the symbol properties of the bounded invertible weighted composition operators. Moreover, we establish the equivalence between bounded invertible operators and frame-preserving operators. Furthermore, we show that weighted composition operator preserves frames if and only if it preserves the Riesz bases property. Additionally,we investigate the weighted composition operators that preserve tight or normalized tight frames on the Dirichlet space.

关键词： Weighted composition operators Weighted Hardy spaces frames invertible

Fine-grained grid computing model for Wi-Fi indoor localization in complex environments

在线全文

学校读者我要写书评

暂无评论

Journal of Electronic Science and Technology 2024年第1期22卷 42-52页

作者： Yan Liang Song Chen Xin dong Tu liu School of Computer Engineering Chengdu Technological UniversityChengdu611730China HAN Networks Corporation Limited Beijing100094China

The fingerprinting-based approach using the wireless local area network(WLAN)is widely used for indoor ***,the construction of the fingerprint database is quite *** when the position of the access point(AP)or wall changes,updating the fingerprint database in real-time is *** appropriate indoor localization approach,which has a low implementation cost,excellent real-time performance,and high localization accuracy and fully considers complex indoor environment factors,is preferred in location-based services(LBSs)*** this paper,we proposed a fine-grained grid computing(FGGC)model to achieve decimeter-level localization *** points(RPs)are generated in the grid by the FGGC ***,the received signal strength(RSS)values at each RP are calculated with the attenuation factors,such as the frequency band,three-dimensional propagation distance,and walls in complex *** a result,the fingerprint database can be established automatically without manual measurement,and the efficiency and cost that the FGGC model takes for the fingerprint database are superior to previous *** proposed indoor localization approach,which estimates the position step by step from the approximate grid location to the fine-grained location,can achieve higher real-time performance and localization accuracy *** mean error of the proposed model is 0.36 m,far lower than that of previous ***,the proposed model is feasible to improve the efficiency and accuracy of Wi-Fi indoor *** also shows high-accuracy performance with a fast running speed even under a large-size *** results indicate that the proposed method can also be suitable for precise marketing,indoor navigation,and emergency rescue.

关键词： Fine-grained grid computing (FGGC) Indoor localization Path loss Random forest Reference points(RPs)

Comprehensive characterization and detection of nut allergens in bakery foods using Q-TOF mass spectrometry and bioinformatics

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Food Quality and Safety 2024年第1期8卷 145-157页

作者： Daokun Xu Haolun Huang zhen liu YumeiWang Qinan liu Xing Jiang JunYang Rui Ling Nanjing Institute for Food and Drug Control NanjingChina State Key Laboratory of Food Science and Technology School of Food Science and TechnologyJiangnan UniversityWuxiChina Demonstration Collaborative Laboratory of Analysis and Detection Technology for Food and Drug Safety Nanjing Institute for Food and Drug Control and Agilent Technologies(China)NanjingChina Collaborative Laboratory for Food Safety Nanjing Institute for Food and Drug Control and SCIEX(China)NanjingChina

Food allergy is a growing health issue worldwide and the demand for sensitive,robust and high-throughput analytical methods is *** recent years,mass spectrometry-based methods have been established for multiple food allergen *** the present study,a novel method was developed for the simultaneous detection of almond,cashew,peanut,and walnut allergens in bakery foods using liquid chromatography–mass *** unique to these four ingredients were extracted,followed by trypsin digestion,quadrupole time-of-flight(Q-TOF)mass spectrometry and bioinformatics *** raw data were processed by de-novo sequencing module plus PEAKS DB(database search)module of the PEAKS software to screen peptides specific to each nut *** thermal stability and uniqueness of these candidate peptides were further verified using triple quadrupole mass spectrometry(QQQ-MS)in multiple reaction monitoring(MRM)*** nut species was represented by four peptides,all of which were validated for label-free quantification(LFQ).Calibration curves were constructed with good linearity and correlation coefficient(r 2)greater than *** limits of detection(LODs)were determined to range from 0.11 to 0.31 mg/kg,and were compared with the reference doses proposed by Voluntary Incidental Trace Allergen Labelling(VITAL).The recoveries of the developed method in incurred bakery food matrices ranged from 72.5%to 92.1%with relative standard deviations(RSD)of<5.2%.The detection of undeclared allergens in commercial bakery food samples confirmed the presence of these *** conclusion,this method provides insight into the qualitative and quantitative detection of trace levels of nut allergens in bakery foods.

关键词： Nut allergen Q-TOF mass spectrometry peptides label-free quantitation bakery foods.