检索结果-内蒙古大学图书馆

6th IEEE International Conference on Pattern Recognition and Artificial Intelligence, PRAI 2023

作者： Zhou, Yao Qin, Yusong Wang, Song Huang, Zengxi Zhang, Dazhi Xihua University School of Computer Science and Software Engineering Chengdu China Xihua University School of Computer and Software Engineering and the Sichuan Xihua Jiaotong Forensics Center Chengdu China Sichuan Police College Criminal Science and Technology Department Luzhou China

ISBN: (纸本)9798350325485

Person Re-Identification (ReID) is a crucial task in computer vision that aims to match persons captured from non-overlapping camera views. In this paper, to alleviate the impacts of background and occlusion, we propose to use instance segmentation and pose estimation methods to create masks for global feature extraction. Furthermore, we divide the pedestrian images into three regions according to pedestrian keypoints, trying to eliminate the alignment errors. This part-based matching strategy also helps to address the occlusion issue. Overall, we construct a deep learning network with three branches, including two global branches and a part-based branch. The two global branches extract global features using segmentation-based mask and the mask derived from pedestrian keypoint heatmaps, respectively. In the end, a weighted fusion strategy is used to combine the global scores and part-based scores for final classification. This network enables us to acquire robust global feature of pedestrians by excluding background and occlusion, and simultaneously address the alignment errors to some extent. Experimental results on these three widely used datasets demonstrate the effectiveness of our method: Specifically, it achieves 64.5% rank-1 accuracy and 54.3% mAP on Occluded-Duke, and 94.4% and 87.1% rank-1 accuracies on Market-1501 and DukeMTMC-reID, respectively. © 2023 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Revisiting Kernel Attention with Correlated Gaussian Process Representation 40

Revisiting Kernel Attention with Correlated Gaussian Process...

引用

40th Conference on Uncertainty in Artificial Intelligence, UAI 2024

作者： Bui, Long Minh Huu, Tho Tran Dinh, Duy Nguyen, Tan Minh Hoang, Trong Nghia FPT Software AI Center Department of Mathematics National University of Singapore Singapore School of Electrical Engineering and Computer Science Washington State University United States

Transformers have increasingly become the de facto method to model sequential data with state-of-the-art performance. Due to its widespread use, being able to estimate and calibrate its modeling uncertainty is important to understand and design robust transformer models. To achieve this, previous works have used Gaussian processes (GPs) to perform uncertainty calibration for the attention units of transformers and attained notable successes. However, such approaches have to confine the transformers to the space of symmetric attention to ensure the necessary symmetric requirement of their GP’s kernel specification, which reduces the representation capacity of the model. To mitigate this restriction, we propose the Correlated Gaussian Process Transformer (CGPT), a new class of transformers whose self-attention units are modeled as cross-covariance between two correlated GPs (CGPs). This allows asymmetries in attention and can enhance the representation capacity of GP-based transformers. We also derive a sparse approximation for CGP to make it scale better. Our empirical studies show that both CGP-based and sparse CGP-based transformers achieve better performance than state-of-the-art GP-based transformers on a variety of benchmark tasks. © 2024 Proceedings of Machine Learning Research. All rights reserved.

关键词： Distribution transformers

来源：评论

学校读者我要写书评

暂无评论

A Method for Abnormal Detection and Poisoned Data Recovery in Clustered Federated Learning 18th

A Method for Abnormal Detection and Poisoned Data Recovery...

引用

18th International Conference on Wireless Artificial Intelligent Computing Systems and Applications, WASA 2024

作者： Chen, Yingying Shi, Lei Xu, Hao Ye, Junyu Xu, Juan School of Computer Science and Information Engineering Hefei University of Technology Hefei230009 China Engineering Research Center of Safety Critical Industrial Measurement and Control Technology Ministry of Education Hefei230009 China

ISBN: (纸本)9783031714665

Clustered Federated Learning (CFL), as a type of Federated Learning (FL) paradigm, divides the clients into multiple clusters through the clustering process and trains them within the clusters, thus improving the overall model training accuracy. But the clustering process is also the weak link, which can lead to the failure of the whole model training if it is interfered or destroyed by the attacker. In this paper, we study the attack methods and defense strategies for attackers to implement data poisoning by tampering with client data, resulting in overall clustering failure. Our defense approach is designed to identify anomalous clients and recover their poisoned data. In order to resist the malicious behavior of attackers, we propose a dual-threshold pixel difference and watermark authentication defense method for detecting the presence of anomalous clients. The method can accurately identify the abnormal fluctuations in the data and effectively screen out the abnormal clients. Meanwhile, we propose a self-embedding watermarking defense algorithm based on shuffling idea for recovering poisoned data. Among them, our self-embedding watermarking algorithm not only accurately locates and recovers the tampered region of the image, but also improves the watermark security to prevent attackers from cracking it easily. Simulation results show that our proposed algorithm can accurately identify the abnormal client and recover the data. Even if the client’s data is massively tampered with, we can recover the tampered data images with high quality and ensure that the CFL clustering is not corrupted. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Asymptotic Performance Analysis of Large-Scale Active IRS-Aided Wireless Network

IEEE Open Journal of the Communications Society

引用

IEEE Open Journal of the Communications Society 2023年 4卷 2684-2696页

作者： Wang, Yan Shu, Feng Zhuang, Zhihong Dong, Rongen Zhang, Qi Wu, Di Wang, Xuehui Yang, Liang Wang, Jiangzhou Hainan University School of Information and Communication Engineering Haikou570228 China Nanjing University of Science and Technology School of Electronic and Optical Engineering Nanjing210094 China Hainan University Communication Engineering and Collaborative Innovation Center of Information Technology Haikou570228 China Hunan University College of Computer Science and Electronic Engineering Changsha410082 China University of Kent School of Engineering CanterburyCT2 7NT United Kingdom

In this paper, the dominant factor affecting the performance of active intelligent reflecting surface (IRS) aided wireless communication networks in Rayleigh fading channel, namely the average signal-to-noise ratio (SNR) γ 0 at IRS, is studied. Making use of the weak law of large numbers, the simple asymptotic expressions for the received SNR at user and the average SNR at IRS are derived as the number N of IRS elements goes to medium-scale and large-scale. When N tends to large-scale, the asymptotic received SNR at user is proved to be a linear increasing function of a product of γ 0 and N. Subsequently, when the base station (BS) transmit power is fixed, there exists an optimal limited reflective power at IRS. At this point, more IRS reflect power will degrade the SNR performance. Additionally, under the total power sum constraint of the BS transmit power and the power reflected by the IRS, an optimal power allocation (PA) strategy is derived and shown to achieve 0.83 bit rate gain over equal PA. Finally, an IRS with finite phase shifters being taken into account, generates phase quantization errors, and further leads to a degradation of receive performance. The corresponding closed-form performance loss expressions for user's asymptotic SNR, achievable rate (AR), and bit error rate (BER) are derived for active IRS. Numerical simulation results show that a 3-bit discrete phase shifter is required to achieve a trivial performance loss for a large-scale active IRS. © 2020 IEEE.

关键词： Wireless networks

来源：评论

学校读者我要写书评

暂无评论

Deep Convolutional Neural Network based Wheat Leaf Disease Detection in Precision Agriculture 5

Deep Convolutional Neural Network based Wheat Leaf Disease D...

引用

5th International Conference on Smart Electronics and Communication, ICOSEC 2024

作者： Dey, Sreyashi Naidu, Roshan Dey, Om Subrato Balasundaram, A. Shaik, Ayesha Vellore Institute of Technology Chennai School of Computer Science and Engineering Chennai India Vellore Institute of Technology Chennai Center for Cyber Physical Systems Chennai India

ISBN: (纸本)9798331504403

Wheat is one of the fundamental crops that sustains billions of people globally. Unfortunately its production faces major challenges from numerous plant based diseases which can drastically decrease yields as well as the economic returns. There exist a variety of frameworks utilising computer vision and deep learning techniques which can be invaluable for numerous tasks aiding the stakeholders in agriculture and crop management by identifying diseased sections of *** methods to detect the disease are prone to human errors and such limitations underscore the need for automated and reliable disease detection systems. Also AI is likely to predict a future outbreak of any such disease by analysing the historical and geographical data. In this paper, a novel technique have been proposed whose novelty lies in the result that has been achieved using the proposed algorithm which when compared to two other deep learning models that are ResNet50 and VGG16 achieved a higher accuracy. The deep learning approach for wheat leaf disease detection implemented in this paper is performed using a convolutional neural network (CNN) architecture implemented in Python, OpenCV, NumPy, and TensorFlow alongside implementing some data augmentation leveraging the power of deep learning to extract complex features from wheat leaf images enabling accurate classification between healthy and diseased leaves. The aforementioned algorithm has been trained using a default Crop Disease Dataset extracted from a Github repository which has been able to achieve an accuracy of 88.53% with one full cycle of 30 epochs recording its final epoch with 91.04% accuracy and the final K Fold procedure resulting in 91.42% accuracy. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Ethereum Smart Contracts Vulnerabilities Detection Leveraging Fine-Tuning DistilBERT 10

Ethereum Smart Contracts Vulnerabilities Detection Leveragin...

引用

10th International Conference on Platform Technology and Service, PlatCon 2024

作者： Kim, Jaehyun Le, Thi-Thu-Huong Lee, Sangmyeong Kim, Howon School of Computer Science and Engineering Pusan National University Busan Korea Republic of Blockchain Platform Research Center Pusan National University Busan Korea Republic of

ISBN: (纸本)9798350367874

In recent years, the proliferation of blockchain technology has led to the widespread adoption of smart contracts, particularly on the Ethereum platform. However, these smart contracts are prone to vulnerabilities that can be exploited, leading to significant financial losses and security breaches. This paper explores the application of Distil-BERT, a distilled version of the BERT (Bidirectional Encoder Representations from Transformers) model, to detect vulnerabilities in Ethereum smart contracts. By fine-tuning Distil-BERT on the SB-curate dataset, a curated dataset of vulnerable Solidity smart contracts, we achieve a performance accuracy exceeding 97%. We provide natural language processing (NLP) techniques that can be leveraged to enhance the accuracy and efficiency of these methods. Our experiments show that Distil-BERT can effectively identify common vulnerabilities, such as reentrancy and integer overflows, outperforming traditional static analysis tools. This research contributes to developing more secure smart contract systems and offers insights into integrating advanced machine learning models in blockchain security. © 2024 IEEE.

关键词： Ethereum

来源：评论

学校读者我要写书评

暂无评论

Accelerated learning and co-optimization of elastocaloric effect and stress hysteresis of elastocaloric alloys

引用

Rare Metals 2024年第12期43卷 6606-6624页

作者： Shi-Yu He Fei Xiao Rui-Hang Hou Shun-Gui Zuo Ying Zhou Xiao-Rong Cai Zhu Li Yan-Ming Wang Aysu Catal-Isik Enrique Galindo-Nava Xue-Jun Jin State Key Lab of Metal Matrix Composite School of Materials Science and EngineeringShanghai Jiao Tong UniversityShanghai200240China Materials Genome Initiative Center Shanghai Jiao Tong UniversityShanghai200240China Institute of Medical Robotics Shanghai Jiao Tong UniversityShanghai200240China Department of Materials Science and Engineering Graduate School of EngineeringOsaka UniversityOsaka565-0871Japan University of Michigan-Shanghai Jiao Tong University Joint Institute Shanghai Jiao Tong UniversityShanghai200240China Department of Mechanical Engineering University College LondonTorrington PlaceLondonWC1E 7JEUK

The development of high-performance advanced elastocaloric alloys is crucial for the implementation of elastocaloric ***,we present a machine learning(ML)framework to accelerate the development of novel elastocaloric alloys with large adiabatic temperature change(ΔT_(ad))and low stress hysteresis(Δσ_(hy)).The comprehensive framework comprises database construction,feature selection,model construction,alloy design and validation,and model *** are selected according to the physical attributes they *** that may reflect the compatibility between parent and product phases,lattice distortion and the free energy in the alloy are considered in the *** them,the key features are selected by recursive feature elimination and exhaustive search *** trained models in combination with the Bayesian optimization method are exploited to achieve multi-objective *** to the results,a newly designed elastocaloric alloy shows a large adiabatic temperature change of 15.2 K and low average stress hysteresis of70.3 MPa at room temperature,which is consistent with our *** predictions of our ML model are interpreted by the Shapley additive exPlainations(SHAP)approach,which explicitly quantifies the effects of each feature in our model on the adiabatic temperature change and stress ***,we employ the Sureindependence screening and sparsifying operator(SISSO)method in conjunction with the key features to formulate explicit model.

关键词： Elastocaloric alloys Adiabatic temperature change Stress hysteresis Machine learning Multi-objective optimization

来源：评论

学校读者我要写书评

暂无评论

Pedestrian Detection in Foggy Weather Through YOLOv8 Based on FEAttention 1

引用

41st computer Graphics International Conference, CGI 2024

作者： Sun, Meng Zhu, Jianlin Yang, Bo Huang, Jin Zhang, Xiao College of Computer Science South-Central Minzu University Wuhan430074 China Hubei Provincial Engineering Research Center for Intelligent Management of Manufacturing Enterprises South-Central Minzu University Wuhan430074 China School of Computer Science and Technology University of Science and Technology of China Hefei230026 China School of Computer Science and Artificial Intelligence Wuhan Textile University Wuhan430200 China

ISBN: (数字)9783031820243

ISBN: (纸本)9783031820236

Despite the high accuracy achieved by current pedestrian detection technologies, their performance is significantly hindered in foggy weather conditions due to image color distortion and an increased false alarm. This often leads to false positives, false negative, and other related issues. To address these limitations, we propose an enhanced FDW-YOLOv8 (Feature-Enhancing-Attention Dark Channel Prior Wise-IOU YOLOv8) model based on the YOLOv8 algorithm. This model incorporates FEAttention(Feature-Enhancing-Attention) mechanism into the original YOLOv8 network to improve the discriminability and robustness of extracted features, thus reducing the miss rate(MR). Furthermore, we employ the WiseIOU(WIoU) loss function to expedite model convergence and optimize the accuracy of anchor frames, effectively handling challenges posed by occlusions and scale variations in foggy environments. Prior to model prediction, we utilize the Dark Channel Prior(DCP) defogging algorithm to preprocess input images, effectively extracting pedestrian information while suppressing irrelevant details, thus minimizing the false positive rate (FPR). To evaluate the performance of our model, we create a specialized Foggy-Pedestrian dataset, focusing on pedestrian detection in foggy conditions. Our experiments demonstrate that the proposed FDW-YOLOv8 model significantly outperforms the benchmark model in terms of reducing FPR and MR, exhibiting high generalization and practical applicability for real-world applications in foggy environments. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

Inaudible Sound Covert Channel with Anti-Jamming Capability: Attacks vs. Countermeasure

引用

Journal of Electronic science and Technology 2022年第4期20卷 396-415页

作者： Xiao-Hang Wang Shou-Bin Li Ying-Tao Jiang Amit Kumar Singh Bi-Yun Ma Le-Tian Huang Mei Yang Fen Guo School of Cyber Science and Technology Zhejiang UniversityHangzhou 310012 ZJU-Hangzhou Global Scientific and Technological Innovation Center Hangzhou 310012 School of Software Engineering South China University of TechnologyGuangzhou 510006 Department of Electrical and Computer Engineering University of NevadaLas Vegas 89154 School of Computer Science and Electronic Engineering University of EssexColchester CO43SQ School of Electronics and Information Engineering South China University of TechnologyGuangzhou 510640 School of Electronic Science and Engineering University of Electronic Science and Technology of ChinaChengdu 610054

When an inaudible sound covert channel(ISCC)attack is launched inside a computer system,sensitive data are converted to inaudible sound waves and then *** receiver at the other end picks up the sound signal,from which the original sensitive data can be *** a forceful countermeasure against the ISCC attack,strong noise can be used to jam the channel and literally shut down any possible sound data *** this paper,enhanced ISCC is proposed,whose transmission frequency can be dynamically ***,if the transmitter detects that the covert channel is being jammed,the transmitter and receiver both will switch to another available frequency and re-establish their communications,following the proposed communications *** results show that the proposed enhanced ISCC can remain connected even in the presence of a strong jamming noise ***,a detection method based on frequency scanning is proposed to help to combat such an anti-jamming sound *** the proposed countermeasure,the bit error rate(BER)of the data communications over enhanced ISCC soars to more than 48%,essentially shutting down the data transmission,and thus neutralizing the security threat.

关键词： Countermeasure inaudible sound covert channel(ISCC)

来源：评论

学校读者我要写书评

暂无评论

PSST: A Benchmark for Evaluation-driven Text Public-Speaking Style Transfer

PSST: A Benchmark for Evaluation-driven Text Public-Speaking...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Sun, Huashan Wu, Yixiao Ye, Yuhao Yang, Yizhe Li, Yinghao Li, Jiawei Gao, Yang School of Computer Science and Technology Beijing Institute of Technology Beijing Engineering Research Center of High Volume Language Information Processing and Cloud Computing Applications China

ISBN: (纸本)9798891761681

Language style is necessary for AI systems to understand and generate diverse human language ***, previous text style transfer primarily focused on sentence-level data-driven approaches, limiting exploration of potential problems in large language models (LLMs) and the ability to meet complex application *** overcome these limitations, we introduce a novel task called Public-Speaking Style Transfer (PSST), which aims to simulate humans to transform passage-level, official texts into a public-speaking *** in the analysis of real-world data from a linguistic perspective, we decompose public-speaking style into key sub-styles to pose challenges and quantify the style modeling capability of *** such intricate text style transfer, we further propose a fine-grained evaluation framework to analyze the characteristics and identify the problems of stylized *** experiments suggest that current LLMs struggle to generate public speaking texts that align with human preferences, primarily due to excessive stylization and loss of semantic information. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

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：