检索结果-内蒙古大学图书馆

24th International Conference on Web Information Systems engineering, WISE 2023

作者： He, Kun Meng, Shunmei Li, Qianmu Liu, Xiao Beheshti, Amin Chi, Xiaoxiao Zhang, Xuyun School of Cyber Science and Engineering Nanjing University of Science and Technology Nanjing China School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China State Key Laboratory for Novel Software Technology Nanjing University Nanjing China School of Computing Macquarie University Sydney Australia

ISBN: (纸本)9789819972531

Recently, contrastive learning has been widely used in the field of sequential recommendation to solve the data sparsity problem. CL4Rec augments data through simple random crop, mask, and reorder, while DuoRec proposes a model-level data augmentation method. However, these methods do not take into account the issue of noisy data in sequential recommendation, such as false clicks during browsing. The noise may lead to poor representations of learned sequences and negatively affect the augmented data. Current sequential recommendation methods tend to learn the user’s intention from their original sequences, but these methods have certain limitations as the user’s intention for the next interaction may change. Based on the above observations, we propose Noise-augmented Contrastive Learning for Sequential Recommendation (NCL4Rec). Our NCL4Rec proposes sequential noise probability-guided data augmentation. We introduce supervised noise recognition during training instead of obtaining it from original sequences. Moreover, we design positive and negative augmentations of the sequence and design unique noise loss function to train them. Through experiments, it is verified that our NCL4Rec consistently outperforms the current state-of-the-art models. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Contrastive Learning Sequential Recommendation

来源：评论

学校读者我要写书评

暂无评论

Expert with Outlier Exposure for Continual Learning of New Diseases

Expert with Outlier Exposure for Continual Learning of New D...

引用

2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022

作者： Xu, Zhengjing Chen, Kanghao Zheng, Wei-Shi Tan, Zhijun Yang, Xiaobo Wang, Ruixuan Sun Yat-sen University School of Computer Science and Engineering Guangzhou China Key Laboratory of Machine Intelligence and Advanced Computing MOE China State Key Laboratory of Dampness Syndrome of Chinese Medicine Guangzhou China Peng Cheng Laboratory Department of Network Intelligence Shenzhen China

ISBN: (纸本)9781665468190

Current intelligent diagnosis systems struggle to continually learn to diagnose more and more diseases due to catastrophic forgetting of old knowledge when learning new knowledge. Although storing small old data for subsequent continual learning can effectively help alleviate the forgetting issue, the heavy data imbalance between old classes and to-be-learned new classes in classifier training often causes biased prediction towards the new classes just learned by the updated classifier. In this study, an outlier detection technique is novelly applied to train an additional expert classifier for new classes to help alleviate the class imbalance issue and discriminate the learned new classes from old classes during inference (instead of the training phase). Specially, the stored small data of old classes are considered as outliers during training the expert classifier, such that the output probability distributions from the expert classifier are expected to be obviously different between test data of the old classes and those of the new classes. Such difference between old classes and new classes can be used to fine-tune the original output from the updated classifier which is responsible for prediction of all learned (old and new) classes. During inference, a novel ensemble strategy is proposed to combine the predictions from the updated classifier, the expert classifier, and the previously learned old classifier. The proposed learning and inference framework can be easily combined with existing continual learning strategies. Empirical evaluations on three medical image datasets and one natural image dataset show that the proposed framework can effectively improve continual learning performance. © 2022 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Adversarial Example Attacks Against Intelligent Malware Detection: A Survey

Adversarial Example Attacks Against Intelligent Malware Dete...

引用

International Conference on Applied Machine Learning (ICAML)

作者： Xuyan Qi Yonghe Tang Huanwei Wang Tieming Liu Jing Jing State Key Laboratory of Mathematic Engineering and Advanced Computing Zhengzhou China

With the advancement of information technology and the increasing prevalence of the internet, the number of malware has experienced an exponential, and the security threat to users' property and privacy is becoming more serious. With the great success of artificial intelligence especially deep learning technology in fields such as computer vision and speech processing, deep learning are gradually introduced into the field of malware detection, to process large amounts of malware more efficiently. However, in recent years, researchers have discovered that the deep learning model itself also has security risks. While malware detection based on deep learning brings convenience and technical progress, it also introduces some new threats, such as adversarial examples. This paper investigates adversarial example attacks against deep learning-based malware detectors. First, We reviewed research on malware detection techniques. Then we analyze the works related to adversarial example attacks against intelligent malware detectors. Finally, we discuss the challenges and prospects in future research.

关键词： Deep learning Computer vision Technological innovation Privacy Perturbation methods Detectors Predictive models

来源：评论

学校读者我要写书评

暂无评论

Traffic Modeling and Rescheduling for High-speed Train Based on Block Sections 4

Traffic Modeling and Rescheduling for High-speed Train Based...

引用

4th International Conference on Data-Driven Optimization of Complex Systems, DOCS 2022

作者： Yue, Peng Jin, Yaochu Dai, Xuewu Cui, Dongliang Shi, Qi Northeastern University State Key Laboratory of Synthetical Automation for Process Industries Shenyang110819 China University of Bielefeld Nature Inspired Computing and Engineering Group BielefeldD-33619 Germany

ISBN: (数字)9781665459822

ISBN: (纸本)9781665459822

Affected by unexpected events, the nominal operation of high-speed trains will become invalid. To maintain the efficiency of trains, train dispatchers need to reschedule the train timetable, which is a challenging task. On the one hand, the dispatchers need to take into account complex conflicts between trains on the track;on the other hand, the rescheduled timetable should be efficient to reduce operating costs. To address the above issues, this study proposes a traffic modeling method for high-speed trains based on a block section to describe in detail the operation conflicts between trains. A train rescheduling approach combining reinforcement learning and model predictive control is proposed to accomplish train rescheduling efficiently. The experiments show the effectiveness of the proposed method. © 2022 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

MDFA: A Malware Detection Framework Based on Attention Mechanism with Bi-LSTM

MDFA: A Malware Detection Framework Based on Attention Mecha...

引用

International Conference on Applied Machine Learning (ICAML)

作者： Xuyan Qi Huanwei Wang Yonghe Tang Wei Liu Jiangtao Xie State Key Laboratory of Mathematic Engineering and Advanced Computing Zhengzhou China

As the existing malware intelligent detection methods have shortcomings and low accuracy of manual feature extraction and feature processing, a Malware Detection Framework with Attention mechanism based on Bi-directional Long Short-Term Memory (MDFA) was proposed. To training this method, We have collected 15, 893 malware and 10, 959 benign program samples and constructed a dataset. Then, the sample data was labeled and sanitized by determining the results based on VirusTotal analysis. Finally, the Bi-directional Long Short-Term Memory model injected into the attention mechanism was built and trained to complete the malware detection. Experimental results show that the accuracy of MDFA was improved by at least 2. 3%, and the recall rate was improved significantly compared with the existing methods, verifying the effectiveness of MDFA.

关键词： Training Computer viruses Bidirectional control Manuals Machine learning Feature extraction Transformers

来源：评论

学校读者我要写书评

暂无评论

Excellent Student Training in Shielding Effectiveness Analysis with International Cooperation 10

Excellent Student Training in Shielding Effectiveness Analys...

引用

10th IEEE International Conference on Computer Science and Network Technology, ICCSNT 2022

作者： Li, Yingsong Huang, Zhixiang Wang, Meiyu Jiang, Tao Sinkevich, Eugene Mordachev, Vladimir College of Information and Communication Engineering Harbin Engineering University Harbin15001 China Anhui University Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education China Belarusian State University of Informatics and Radioelectronics Emc RandD Laboratory Minsk Belarus

ISBN: (数字)9781665470612

ISBN: (纸本)9781665470612

We present the excellent student training scheme in shielding effectiveness analysis with international cooperation of Belarus. As we know, the shielding is vital for modern information devices due to the radiating around us. Herein, the shielding effectiveness is important for training excellent students to master the design of the frequency selective components. In this method, we teach the students to use the EM simulation tools to construct different EM structures for obtaining the shielding, and the performance of these structures are verified from the simulations and measurements. Using this shielding effectiveness training scheme, a frequency selective surface is presented, simulated and discussed to analyze its performance. In future, the excellent student training methods given herein for shielding effectiveness can be expanded to develop other EM structures. © 2022 IEEE.

关键词： International cooperation

来源：评论

学校读者我要写书评

暂无评论

A Neighbor Immunization Strategy to Block Virus Spreading on Networks 2

A Neighbor Immunization Strategy to Block Virus Spreading on...

引用

2nd Information Communication Technologies Conference, ICTC 2021

作者： Yaohui, Hao Yingze, Liu Baole, Zhao Qingxian, Wang Information Engineering University State Key Laboratory of Mathematical Engineering and Advanced Computing Zhengzhou China Information Engineering University Zhengzhou China

ISBN: (纸本)9780738142876

In order to achieve better effect on blocking virus propagation, a neighbor immunization strategy was proposed and verified in this paper. On email-EU real networks and on scale-free model network, the efficiencies of the neighbor immunization strategy and the random immunization strategy in SIR and SEIR virus spreading models were compared and analyzed by simulation experiment. The result indicates that the neighbor immunization strategy is a feasible and effective. This phenomenon shows that neighbor immunization strategy can improve to the future virus and effectively prevent the virus spreading. © 2021 IEEE.

关键词： Immunization

来源：评论

学校读者我要写书评

暂无评论

Pruned-F1DCN: A lightweight network model for traffic classification 22

Pruned-F1DCN: A lightweight network model for traffic classi...

引用

Proceedings of the 2022 International Conference on Cyber Security

作者： Ruo nan Wang Jin long Fei Rong kai Zhang State Key Laboratory of Mathematical Engineering and Advanced Computing PLA Strategic Support Force Information Engineering University Zhengzhou China China

ISBN: (纸本)9798400700132

With the continuous development of deep learning, deep neural networks are gradually applied to traffic classification problems. However, the large network structure and parameter number of deep neural networks hinder the application on edge computing devices. Reducing network scale helps relieve computational pressure, this paper proposes a lightweight traffic classification model to provide reliable accuracy and reduce the consumption of computing resources. In this work, we design an F1DCN network, which takes full advantage of the convolution layer parameters and the convolution kernel field of view. The lightweight approach effectively improves the classification performance and saves massive parameters. The model pruning method is applied to find the optimal structure of the network. Experiments on two public datasets show that the proposed model reduce more than 80 % parameters and 45 % FLOPS compared with traditional traffic classification methods, and maintaining more than 95 % classification accuracy.

关键词： Convolutional neural network Lightweight Model pruning Traffic classification

来源：评论

学校读者我要写书评

暂无评论

Health risk ranking of antibiotic resistance genes in the Yangtze River

引用

Environmental Science and Ecotechnology 2024年第5期21卷 113-122页

作者： Chunxia Jiang Zelong Zhao Hans-Peter Grossart Feng Ju Yi Zhao Geoffrey Michael Gadd Ewa Korzeniewska Yuyi Yang Key Laboratory of Aquatic Botany and Watershed Ecology Wuhan Botanical GardenChinese Academy of SciencesWuhan430074China University of Chinese Academy of Sciences Beijing100049China Danjiangkou Wetland Ecosystem Field Scientific Observation and Research Station Chinese Academy of Sciences&Hubei ProvinceWuhan430074China Liaoning Key Laboratory of Marine Fishery Molecular Biology Liaoning Key Lab of Germplasm Improvement and Fine Seed Breeding of Marine Aquatic AnimalsLiaoning Ocean and Fisheries Science Research InstituteDalian116023China Leibniz-Institute for Freshwater Ecology and Inland Fisheries(IGB) Neuglobsow16775Germany Institute for Biochemistry and Biology Potsdam UniversityPotsdam14469Germany Key Laboratory of Coastal Environment and Resources of Zhejiang Province School of EngineeringWestlake UniversityHangzhou310030China School of Water Resources and Environment China University of Geosciences(Beijing)Beijing100083China Geomicrobiology Group School of Life SciencesUniversity of DundeeDundeeDD15EHScotlandUK State Key Laboratory of Heavy Oil Processing State Key Laboratory of Petroleum Pollution ControlChina University of PetroleumBeijing102249China Department of Water Protection Engineering and Environmental Microbiology Faculty of GeoengineeringUniversity of Warmia and Mazury in OlsztynOlsztyn10-720Poland

Antibiotic resistance is an escalating global health concern,exacerbated by the pervasive presence of antibiotic resistance genes(ARGs)in natural *** Yangtze River,the world's third-longest river,traversing areas with intense human activities,presents a unique ecosystem for studying the impact of these genes on human ***,we explored ARGs in the Yangtze River,examining 204 samples from six distinct habitats of approximately 6000 km of the river,including free-living and particle-associated settings,surface and bottom sediments,and surface and bottom bank *** shotgun sequencing,we generated an average of 13.69 Gb reads per *** findings revealed a significantly higher abundance and diversity of ARGs in water-borne bacteria compared to other habitats.A notable pattern of resistome coalescence was observed within similar habitat *** addition,we developed a framework for ranking the risk of ARG and a corresponding method for calculating the risk *** them,we identified water-borne bacteria as the highest contributors to health risks,and noted an increase in ARG risks in particle-associated bacteria correlating with heightened anthropogenic *** analysis using a weighted ARG risk index pinpointed the ChengdueChongqing and Yangtze River Delta urban agglomerations as regions of elevated health *** insights provide a critical new perspective on ARG health risk assessment,highlighting the urgent need for strategies to mitigate the impact of ARGs on human health and to preserve the ecological and economic sustainability of the Yangtze River for future human use.

关键词： Antibiotic resistance genes Yangtze River Habitat Coalescence Health risk

来源：评论

学校读者我要写书评

暂无评论

Asymptotic Estimates for Spectral Estimators of Rotationally Invariant Matrices

Asymptotic Estimates for Spectral Estimators of Rotationally...

引用

IEEE International Symposium on Information Theory

作者： Zhuohang He Xiaojun Yuan Junjie Ma National Key Laboratory of Wireless Communications University of Electronic Science and Technology of China Chengdu China Institute of Computational Mathematics and Scientific/Engineering Computing AMSS Chinese Academy of Sciences Beijing China

ISBN: (数字)9798350382846

ISBN: (纸本)9798350382853

In this paper, we consider the recovery of low-rank matrices from noisy observations using spectral denoisers, where the singular values are denoised through an identical scalar smoothing function. We explore the asymptotic mean squared error (AMSE) of these denoisers within a framework where the rank of the matrix to be recovered grows linearly with the matrix size. We demonstrate that, under arbitrary i.i.d. noise and some mild regularity assumptions, the AMSE converges in probability to a deterministic function of the noise power. Our results are applicable to commonly used denoisers, including the best-rank-r denoiser, the singular-value soft-threshold denoiser, and the singular-value hard-threshold denoiser. To the best of our knowledge, this is the first study to establish an analytical expression for the asymptotic MSE under arbitrary i.i.d. noise. The derived analytical expression depends solely on the empirical distribution of the singular values of the low-rank matrix and the specific form of the spectral denoiser employed.

关键词： Smoothing methods Limiting Noise Noise reduction Noise measurement Information theory

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：