检索结果-内蒙古大学图书馆

Boost AI power: Data augmentation strategies with unlabelled data and conformal prediction, a case in alternative herbal medicine discrimination with electronic nose

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Liu, Li Zhan, Xianghao Wu, Rumeng Guan, Xiaoqing Wang, Zhan Zhang, Wei Pilanci, Mert Wang, You Luo, Zhiyuan Li, Guang The State Key Laboratory of Industrial Control Technology Institute of Cyber-Systems and Control Zhejiang University Hangzhou310027 China Department of Bioengineering Stanford University StanfordCA94305 United States The Computer Learning Research Center Royal Holloway University of London Egham Hill Surrey EghamTW20 0EX United Kingdom Department of Electrical Engineering Stanford University StanfordCA94305 United States

Electronic nose has been proven effective in alternative herbal medicine classification, but due to the nature of supervised learning, previous research heavily relies on the labelled training data, which are time-costly and labor-intensive to collect. To alleviate the critical dependency on the training data in real-world applications, this study aims to improve classification accuracy via data augmentation strategies. The effectiveness of five data augmentation strategies under different training data inadequacy are investigated in two scenarios: the noise-free scenario where different availabilities of unlabelled data were considered, and the noisy scenario where different levels of Gaussian noises and translational shift were added to represent sensor drifts. The five augmentation strategies, namely noise-adding data augmentation, semi-supervised learning, classifier-based online learning, Inductive Conformal Prediction (ICP) online learning and our novel ensemble ICP online learning (EICP) proposed in this study, are compared against supervised learning baselines, with Linear Discriminant Analysis (LDA) and Support Vector Machine (SVM) as the classifiers. This study provides a systematic analysis of different augmentation strategies. Our novel strategy, EICP, outperforms the others by showing non-decreasing classification accuracy on all tasks and a significant improvement on most simulated tasks (25 out of 36 tasks, p ≤ 0.05), which demonstrated both effectiveness and robustness in boosting the classification model generalizability. This strategy can be employed in other machine learning applications. Copyright © 2021, The Authors. All rights reserved.

关键词： Electronic nose

A blockchain future for internet of things security： a position paper

学校读者我要写书评

暂无评论

Digital Communications and Networks 2018年第3期4卷 149-160页

作者： Mandrita Banerjee Junghee Lee Kim-Kwang Raymond Choo Department of Electrical and Computer Engineering University of Texas at San Antonio San Antonio TX 78249 USA Department of Information Systems and Cyber Security University of Texas at San Antonio San Antonio TX 78249 USA

Internet of Things （IoT） devices are increasingly being fotmd in civilian and military contexts, ranging from smart cities and smart grids to Internet-of-Medical-Things, Internet-of-Vehicles, Internet-of-Military-Things, Internet-of- Battlefield-Things, etc. In this paper, we survey articles presenting IoT security solutions published in English since January 2016. We make a nmnber of observations, including the lack of publicly available loT datasets that can be used by the research and practitioner communities. Given the potentially sensitive nature of IoT datasets, there is a need to develop a standard for sharing IoT datasets among the research and practitioner communities and other relevant stakeholders. Thus, we posit the potential for blockehain technology in facilitating secure sharing of IoT datasets （e.g., using blockchain to ensure the integrity of shared datasets） and securing loT systems, before presenting two conceptual blockchain-based approaches. We then conclude this paper with nine potential research questions.

关键词： Blockchain Blockchain security Collaborative security Internet-of-military things dataset self-healing security Intrusion-prevention system Predictive security Predictive security

Univariate Forecasting of time series data Based on Seq2Seq Model 20

学校读者我要写书评

暂无评论

Univariate Forecasting of time series data Based on Seq2Seq ...

Proceedings of the 2020 4th International Conference on Electronic Information technology and computer engineering

作者： Yanwen Guo Tao Wu Yanbing Liu Yixin Jiang Department of Computer Science and Technology Chongqing University of Posts and Telecommunications Chongqing China School of Cybersecurity and Information Law Chongqing University of Posts and Telecommunications Chongqing China Chongqing Engineering Laboratory of Internet and Information Sccurity Chongqing University of Posts and Telecommunications Chongqing China Electric Power Research Institute CSG Guangdong Provincial Key Laboratory of Power System Network Security Guangzhou China

ISBN: (纸本)9781450387811

With the release of country's 2020 new infrastructure projects, the increasing demand for electricity in various projects has further promoted the intelligent upgrade of the power grid. The processing of grid data at the edge has become one of the future trends of next generation grid. In the field of power grid edge time series data prediction, the inability to obtain the variable characteristics of other systems and the uncertainty of the length of various edge input data, this paper proposes a Seq2seq prediction model based on a Bi-directional Long Short Term Memory (Bi-LSTM). Bi-LSTM can make full use of contextual information to achieve a more accurate prediction effect. The Seq2Seq model can process variable-length sequence data. In this paper, Seq2Seq model and BI-LSTM model are combined to solve the problem of uncertain edge length and univariate prediction. At the same time, in order to solve the information loss caused by Seq2Seq in the encoding stage, the Attention mechanism is introduced, it can compute the allocation of resources to more important tasks while addressing the problem of information overload. Through experimental research, it is found that the proposed ABSeq2Seq model is superior to other classic models in univariate prediction, and it has the better generalization for various types of data. We have proved its effectiveness in the single variable power prediction.

关键词： Time Series Bi-LSTM Seq2Seq Prediction Attention Mechanism Univariate

Budgeted online selection of candidate iot clients to participate in federated learning

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Mohammed, Ihab Tabatabai, Shadha Al-Fuqaha, Ala El Bouanani, Faissal Qadir, Junaid Qolomany, Basheer Guizani, Mohsen School of Computer Sciences Western Illinois University MacombIL61455 United States Department of Computer Science Western Michigan University and Computer Science Department Al-Nahrain University Iraq Information and Computing Technology Division Hamad Bin Khalifa University and Computer Science Department Western Michigan University United States Mohammed V University Rabat Morocco Information Technology University Lahore Pakistan Department of Cyber Systems College of Business & Technology University of Nebraska at Kearney KearneyNE68849 United States Computer Science and Engineering Department Qatar University Doha Qatar

Machine Learning (ML), and Deep Learning (DL) in particular, play a vital role in providing smart services to the industry. These techniques however suffer from privacy and security concerns since data is collected from clients and then stored and processed at a central location. Federated Learning (FL), an architecture in which model parameters are exchanged instead of client data, has been proposed as a solution to these concerns. Nevertheless, FL trains a global model by communicating with clients over communication rounds, which introduces more traffic on the network and increases the convergence time to the target accuracy. In this work, we solve the problem of optimizing accuracy in stateful FL with a budgeted number of candidate clients by selecting the best candidate clients in terms of test accuracy to participate in the training process. Next, we propose an online stateful FL heuristic to find the best candidate clients. Additionally, we propose an IoT client alarm application that utilizes the proposed heuristic in training a stateful FL global model based on IoT device type classification to alert clients about unauthorized IoT devices in their environment. To test the efficiency of the proposed online heuristic, we conduct several experiments using a real dataset and compare the results against state-of-the-art algorithms. Our results indicate that the proposed heuristic outperforms the online random algorithm with up to 27% gain in accuracy. Additionally, the performance of the proposed online heuristic is comparable to the performance of the best offline algorithm. Copyright © 2020, The Authors. All rights reserved.

关键词： Internet of things

Trading Based Service-Oriented Spectrum-Aware RAN-Slicing Under Spectrum Sharing

学校读者我要写书评

暂无评论

Trading Based Service-Oriented Spectrum-Aware RAN-Slicing Un...

IEEE Conference on Wireless Communications and Networking

作者： Kajia Jiao Xuanheng Li Miao Pan Fan Jiang Ming Li School of Information and Communication Engineering Dalian University of Technology Dalian China Department of Electrical and Computer Engineering University of Houston Houston TX USA Shaanxi Key Laboratory of Information Communication Network and Security Xi’an University of Posts and Telecommunications Xian Shaanxi China

ISBN: (数字)9781728131061

ISBN: (纸本)9781728131078

The fast development on emerging services makes our telecommunications networks witness two key problems. One is the flexibility to fulfill the diverse service requests and the other is the shortage on spectrum. Network slicing and spectrum sharing have been regarded as two prominent solutions, which, however, are barely jointly studied. When taking the shared spectrum into account, its unique feature of heterogeneity and uncertainty will bring new challenges for the slicing. In this paper, we propose a service-oriented spectrum-aware RAN-slicing trading (SSRT) scheme with a comprehensive consideration on both aspects. For the SSRT scheme, we jointly slice three kinds of resources, namely, time, spectrum (including both licensed one and shared one), and network facilities, according to the diverse traffic requests, which are classified into delay-tolerant (DT) ones and delay-sensitive (DS) ones, as well as the willing payments from different service providers (SPs). To achieve both inter-slice and intra-slice isolation, we construct a three-dimensional (3D) conflict graph and formulate the SSRT scheme into a mixed-integer nonlinear programming (MINLP) problem with a cross-layer spectrum-aware resource allocation and a hybrid transmission mode (including both single-hop and multi-hop). Since finding all the maximum independent sets (MIS) for the 3D conflict graph is an NP-hard problem, we further develop an iterative heuristic algorithm for the MIS determination.

关键词：

That’s Not Me! Designing Fictitious Profiles to Answer securityQuestions

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Micallef, Nicholas Gamagedara Arachchilage, Nalin Asanka UNSW Canberra Cyber School of Engineering and Information Technology University of New South Wales Canberra Australia Optus La Trobe Cyber Security Research Hub Department of Computer Science IT School of Engineering and Mathematical Sciences La Trobe University VIC3086 Australia

Although security questions are still widely adopted, they still have several limitations. Previous research found that using system-generated information to answer security questions could be more secure than users’ own answers. However, using system-generated information has usability limitations. To improve usability, previous research proposed the design of system-generated fictitious profiles. The information from these profiles would be used to answer security questions. However, no research has studied the elements that could influence the design of fictitious profiles or systems that use them to answer security questions. To address this research gap, we conducted an empirical investigation through 20 structured interviews. Our main findings revealed that to improve the design of fictitious profiles, users should be given the option to configure the profiles to make them relatable, interesting and memorable. We also found that the security questions currently provided by websites would need to be enhanced to cater for fictitious profiles. Copyright © 2019, The Authors. All rights reserved.

关键词： Information use

Random Sampling on an High-Dimensional Sphere for Solving SVP

学校读者我要写书评

暂无评论

Random Sampling on an High-Dimensional Sphere for Solving SV...

Computing Conference, 2019

作者： Qin, Ling Zhang, Xue Wang, Xiaoyun Zheng, Zhongxiang Qi, Wenfeng National Digital Switching System Engineering & Technological Research Center Zhengzhou China Institute for Advanced Study Tsinghua University Beijing China Department of Computer Science and Technology Tsinghua University Beijing China Science and Technology on Communication Security Laboratory Chengdu China

ISBN: (纸本)9783030228675

In this paper, we present a random sampling algorithm to solve the shortest vector problem (SVP). The idea of our algorithm comes from the Gaussian hypothesis, which is one of the most important and commonly used hypotheses in solving SVP. The main idea of our method is to generate the points which approximately satisfy Gaussian hypothesis of a given lattice. To find the shortest vector, we set some rules to make the result of this sampling algorithm get shorter progressively. We evaluate our algorithm by statistical and geometrical tools and give a basic analysis of the efficiency of algorithm. According to our analysis, parameters can be optimized to achieve high efficiency. Practical experiments have been conducted to validate our result where we solve SVP instances with dimension from 80 to 100. © 2019, Springer Nature Switzerland AG.

关键词： Computational geometry

Edge Learning for B5G Networks with Distributed Signal Processing: Semantic Communication, Edge Computing, and Wireless Sensing

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Xu, Wei Yang, Zhaohui Ng, Derrick Wing Kwan Levorato, Marco Eldar, Yonina C. Debbah, Mérouane The National Mobile Communications Research Lab Frontiers Science Center for Mobile Information Communication and Security Southeast University Nanjing210096 China Purple Mountain Laboratories Nanjing211111 China The Department of Electronic and Electrical Engineering University College London LondonWC1E 6BT United Kingdom The School of Electrical Engineering and Telecommunications The University of New South Wales Australia The Department of Computer Science University of California IrvineCA92697 United States The Faculty of Math and CS Weizmann Institute of Science Rehovot7610001 Israel The Technology Innovation Institute The Mohamed Bin Zayed University of Artificial Intelligence Masdar City Abu Dhabi9639 United Arab Emirates

To process and transfer large amounts of data in emerging wireless services, it has become increasingly appealing to exploit distributed data communication and learning. Specifically, edge learning (EL) enables local model training on geographically disperse edge nodes and minimizes the need for frequent data exchange. However, the current design of separating EL deployment and communication optimization does not yet reap the promised benefits of distributed signal processing, and sometimes suffers from excessive signalling overhead, long processing delay, and unstable learning convergence. In this paper, we provide an overview on practical distributed EL techniques and their interplay with advanced communication optimization designs. In particular, typical performance metrics for dual-functional learning and communication networks are discussed. Also, recent achievements of enabling techniques for the dual-functional design are surveyed with exemplifications from the mutual perspectives of "communications for learning" and "learning for communications." The application of EL techniques within a variety of future communication systems are also envisioned for beyond 5G (B5G) wireless networks. For the application in goal-oriented semantic communication, we present a first mathematical model of the goal-oriented source entropy as an optimization problem. In addition, from the viewpoint of information theory, we identify fundamental open problems of characterizing rate regions for communication networks supporting distributed learning-and-computing tasks. We also present technical challenges as well as emerging application opportunities in this field, with the aim of inspiring future research and promoting widespread developments of EL in B5G. Copyright © 2022, The Authors. All rights reserved.

关键词： Electronic data interchange

An Intelligent Wireless Displacement Sensor for Landslide Monitoring and Early Warning

学校读者我要写书评

暂无评论

IOP Conference Series: Earth and Environmental Science 2021年第7期861卷

作者： Xing Zhu Hewei Xi Zhaoqing He Liu Yang State Key Laboratory Geohazard Prevention and Geoenvironment Protection Chengdu University of Technology Chengdu 610059 China College of Computer Science and Cyber Security Chengdu Univeristy of Technology Chengdu 610059 China College of Nuclear Technology and Automation Engineering Chengdu University of Technology Chengdu 610059 China

Landslide poses a great threat to the safety of local communities, engineering construction and operation. Accurate monitoring and successful early-warning of rapid landslides using conventional techniques with constant and sparse sampling strategy remains a challenge in geo-disaster research and mitigation. Sampling and uploading all sensor data to cloud imposes huge burden on the power supply and radio bandwidth, which greatly reduces the long-term stability and robustness of the wireless sensor in field. With aim to achieve the best powerful sensing ability based on limited power and bandwidth of sensor, we proposed a low-cost and energy efficient wireless crack-meter sensor integrated with intelligent computing architecture for rapid deformation monitoring. A self-adaptive data sampling and uploading algorithm based on the variation of objective physical value was implemented in the embedded microcontroller of sensor node, and only the valuable information derived from sensor data to the cloud server. Experimental results demonstrated that the intelligent sensing approach can effectively collect the crucial data at a high frequency during accelerating deformation of landslide, while only update status at several hours during non-deformation phase. Several landslide monitoring applications have proved that this intelligent sensing can significantly and effectively enhance the success rate and real-time performance of early-warning for rapid landslide.

关键词：