检索结果-内蒙古大学图书馆

22nd IEEE International Conference on Machine Learning and Applications, ICMLA 2023

作者： Khaled, Kacem Dhaouadi, Mouna De Magalhaes, Felipe Gohring Nicolescu, Gabriela Polytechnique Department of Computer Engineering and Software Engineering Montreal Canada University of Montreal Department of Computer Science and Operations Research Canada

ISBN: (纸本)9798350345346

Model stealing attacks have become a serious concern for deep learning models, where an attacker can steal a trained model by querying its black-box API. This can lead to intellectual property theft and other security and privacy risks. The current state-of-the-art defenses against model stealing attacks suggest adding perturbations to the prediction probabilities. However, they suffer from heavy computations and make impracticable assumptions about the adversary. They often require the training of auxiliary models. This can be time-consuming and resource-intensive which hinders the deployment of these defenses in real-world applications. In this paper, we propose a simple yet effective and efficient defense alternative. We introduce a heuristic approach to perturb the output probabilities. The proposed defense can be easily integrated into models without additional training. We show that our defense is effective in defending against three state-of-the-art stealing attacks. We evaluate our approach on large and quantized (i.e., compressed) Convolutional Neural Networks (CNNs) trained on several vision datasets. Our technique outperforms the state-of-the-art defenses with a ×37 faster inference latency without requiring any additional model and with a low impact on the model's performance. We validate that our defense is also effective for quantized CNNs targeting edge devices. © 2023 IEEE.

关键词： Deep learning Model stealing attacks Privacy Quantization Security

来源：评论

学校读者我要写书评

暂无评论

High-Fidelity Diffusion Editor for Zero-Shot Text-Guided Video Editing 24

High-Fidelity Diffusion Editor for Zero-Shot Text-Guided Vid...

引用

24th IEEE International Conference on Data Mining, ICDM 2024

作者： Luo, Yan Zuo, Zhichao Zhang, Zhao Zhao, Zhongqiu Zhang, Haijun Hong, Richang School of Computer Science and Information Engineering Hefei University of Technology Hefei China Yunnan Key Laboratory of Software Engineering Yunan China School of Computer Science Harbin Institute of Technology Shenzhen China

ISBN: (纸本)9798331506681

Text-guided image generative diffusion models achieve fast development on the generation and editing of high-quality images. To extend such success to video editing, some efforts combining image generation with video editing have been made, which however only achieve inferior performance. We attribute it to two challenges: 1) different from the static image generation, it is tricky for dynamic video information to ensure the temporal fidelity of motion consistency across different frames;2) the randomness of the frame generation process makes it hard to continuously retain the similar spatial fidelity for the original detailed features. In this paper, we propose a new high-fidelity diffusion model-based zero-shot text-guided video editing network, called HiFiVEditor, which aims to conduct effective video editing with high fidelity of the original video's detailed and dynamic information. Specifically, we propose a Spatial-Temporal Fidelity Block (STFB) that enables the model to restore the spatial features by enlarging the spatial perceptual field to avoid loss of important information, and capture more dynamic information between different frames by using all frames for preserving temporal consistency to achieve better temporal fidelity. In addition, we introduce Null-Text Embedding to create a soft text embedding to optimize the noise learning process, so that the latent noise can be aligned with the prompt. Furthermore, to tune the video style and render it more realistic, we employ a Prior-Guided Perceptual Loss to constrain the prediction results to avoid deviating from the original video style. Extensive experiments demonstrate the superior video editing capability compared to existing works. © 2024 IEEE.

关键词： Video recording

来源：评论

学校读者我要写书评

暂无评论

MedNetV3: A Deep Learning Model for MRI Brain Tumor Detection with MobileNetV3 and Advanced Data Augmentation 21

MedNetV3: A Deep Learning Model for MRI Brain Tumor Detectio...

引用

21st International computer Conference on Wavelet Active Media Technology and Information Processing, ICCWAMTIP 2024

作者： Tariq, Hussain Ahmed Fengli, Zhang Salim Shakib, Muhamud Md Javed, Iram School of Information and Software Engineering University of Electronic Science and Technology of China China University of Agriculture Department of Computer Science Faisalabad Pakistan

ISBN: (纸本)9798331519254

In this paper, we present a novel deep learning model medical network (MedNetV3) developed for brain tumor detection. It incorporates advanced data augmentation techniques based on the MobileNetV3 architecture. MedNetV3 is developed for higher classification accuracy at the cost of slightly lower computational efficiency and is therefore suitable for clinical use. The proposed model overcomes the class imbalance problem and can make the model robust to the variations of MRI scans included in the extensive data augmentation methods, such as random rotations, flips. The proposed model trained and evaluated using a publicly available MRI dataset. It achieved a high Area Under the Curve (AUC) value of 0.99 for both the Precision-Recall and Receiver Operating Characteristic (ROC) curves. The results show that MedNetV3 is effective in correctly distinguishing brain tumors and therefore has the potential to be used in medical image classification. © 2024 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Multi-Objective Equilibrium Optimizer for Feature Selection in High-Dimensional English Speech Emotion Recognition

引用

computers, Materials & Continua 2024年第2期78卷 1957-1975页

作者： Liya Yue Pei Hu Shu-Chuan Chu Jeng-Shyang Pan Fanli Business School Nanyang Institute of TechnologyNanyang473004China School of Computer and Software Nanyang Institute of TechnologyNanyang473004China College of Computer Science and Engineering Shandong University of Science and TechnologyQingdao266590China Department of Information Management Chaoyang University of TechnologyTaichung413310Taiwan

Speech emotion recognition(SER)uses acoustic analysis to find features for emotion recognition and examines variations in voice that are caused by *** number of features acquired with acoustic analysis is extremely high,so we introduce a hybrid filter-wrapper feature selection algorithm based on an improved equilibrium optimizer for constructing an emotion recognition *** proposed algorithm implements multi-objective emotion recognition with the minimum number of selected features and maximum ***,we use the information gain and Fisher Score to sort the features extracted from ***,we employ a multi-objective ranking method to evaluate these features and assign different importance to *** with high rankings have a large probability of being ***,we propose a repair strategy to address the problem of duplicate solutions in multi-objective feature selection,which can improve the diversity of solutions and avoid falling into local *** random forest and K-nearest neighbor classifiers,four English speech emotion datasets are employed to test the proposed algorithm(MBEO)as well as other multi-objective emotion identification *** results illustrate that it performs well in inverted generational distance,hypervolume,Pareto solutions,and execution time,and MBEO is appropriate for high-dimensional English SER.

关键词： Speech emotion recognition filter-wrapper high-dimensional feature selection equilibrium optimizer multi-objective

来源：评论

学校读者我要写书评

暂无评论

Dual Cross-Stage Partial Learning for Detecting Objects in Dehazed Images 24

Dual Cross-Stage Partial Learning for Detecting Objects in D...

引用

24th IEEE International Conference on Data Mining, ICDM 2024

作者： Zhao, Jinbiao Zhang, Zhao Ren, Jiahuan Zhang, Haijun Zhao, Zhongqiu Wang, Meng School of Computer Science and Information Engineering Hefei University of Technology Hefei China Yunnan Key Laboratory of Software Engineering Yunan China Harbin Institute of Technology Department of Computer Science Shenzhen China

ISBN: (纸本)9798331506681

Performing an object detection task after the restoration of a hazy image, or rather detecting with the network backbone directly, will result in the inclusion of information mixed with dehazing, which tends to interfere with detection performance. To address these issues, we propose a novel framework for detecting objects in dehazed images via Dual Cross-Stage Partial Learning (DCSP). Specifically, we introduce a Cross-Stage Partial (CSP) module for extracting clean feature information after dehazing. Secondly, to enhance data integrity, we employ a skip-input strategy to supplement information related to object detection features that may be lost during the dehazing task, while avoiding the gradient vanishing problem. In addition, CSP is also introduced to facilitate comprehensive learning of multiple feature representations. Finally, to avoid the inclusion of irrelevant dehazing information in detection, we apply a Ground-Truth Flow at detection network (at dark3), for fine feature information calibration. Additionally, we created a synthetic fog dataset to expand the training data for DCSP. Experimental results on both synthetic and real-world datasets demonstrate the effectiveness and accuracy of the proposed method. The code is available at https://***/zhaojinbiao/DCSP. © 2024 IEEE.

关键词： Demulsification

来源：评论

学校读者我要写书评

暂无评论

GSD-YOLO: a Lightweight Framework for Traffic Sign Detection

IAENG International Journal of Computer Science

引用

IAENG International Journal of computer science 2025年第6期52卷 1815-1823页

作者： Jianbang, Ding Tianwei, Shi Zhehao, Zhang Ruiqi, Wang School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan114051 China School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan China

To address the challenges of detecting small traffic signs, low detection accuracy, complex detection network models, and large parameter counts, this paper proposes a lightweight traffic sign detection framework called GSD-YOLO. This framework integrates the GCSconv module, SENetV2 module, and DySample module. Specifically, the Slim-net architecture, formed by combining the GCSconv and VOVGSCP modules, is employed to replace the Conv and C2f modules in the original YOLOv8 neck network. This substitution not only reduces the number of parameters and computational load during training but also achieves a lightweight design. The lightweight SENetV2 module is incorporated into the backbone of the detection model, further lightening the backbone. In the neck network, the DySample module replaces the original upsampling module, enhancing the model’s upsampling capability while reducing the parameter count. Compared to the original YOLOv8 detection network, Grad-CAM visualization shows that traffic signs in the detected regions appear darker and more concentrated. On the CCTSDB dataset, the proposed framework achieves an mAP@0.5 of 95.5%, which is 2.3% higher than the original YOLOv8 (93.3%). The model size is reduced to 4.2MB, which is 32.2% smaller than the original YOLOv8. The GFLOPs are reduced by 35%, and the overall parameter count is decreased by 40%. The lightweight detection framework proposed in this paper effectively reduces model size, simplifies model complexity, decreases parameter count, and improves detection accuracy, achieving significant improvements in the field of traffic sign detection. © (2025), (International Association of Engineers). All rights reserved.

关键词： Road and street markings

来源：评论

学校读者我要写书评

暂无评论

A three-step SEM-Bayesian network approach for predicting the determinants of CloudIoT-based healthcare adoption

引用

International Journal of Information Technology (Singapore) 2024年 1-21页

作者： Altawaiha, Iyad Atan, Rodziah Yaakob, Razali Bin Abdullah, Rusli Bin Hj Department of Software Engineering and Information Systems Faculty of Computer Science and Information Technology Universiti Putra Malaysia (UPM) Serdang 43400 Malaysia Department of Computer Science Faculty of Computer Science and Information Technology Universiti Putra Malaysia (UPM) Serdang 43400 Malaysia

Adopting the CloudIoT-based healthcare paradigm provides various prospects for medical IT and considerably enhances healthcare services. However, compared to the advanced development of CloudIoT-based healthcare systems, their usage is relatively low among healthcare professionals in hospitals. Thus, a study on the healthcare professionals acceptance and adoption of CloudIoT-based healthcare is critical. This study explores factors influencing healthcare professionals adoption of CloudIoT-based healthcare solutions by extending the Unified Theory of Acceptance and Use of Technology 2 model. Data was collected from 300 healthcare professionals in Jordan through an online questionnaire developed using the Google^® form application. The Bayesian Network and Structural Equation Modeling techniques were used to analyze and validate the proposed model. The results revealed that the healthcare professional’s behavioral intention is directly affected by seven factors: performance expectancy, effort expectancy, facilitating conditions, habit, trust, privacy, and security. The results also indicate that trust mediates the influence of privacy, facilitating conditions, and performance expectancy on behavioral intention. The research results will aid CloudIoT service providers, healthcare organizations, designers, and developers by offering comprehensive knowledge of the significant factors of CloudIoT-based healthcare adoption. A better understanding of these factors will help these stakeholder groups enhance their services and speed up the adoption of CloudIoT in the healthcare area. © The Author(s), under exclusive licence to Bharati Vidyapeeth's Institute of computer Applications and Management 2024.

关键词： Adoption Bayesian network CloudIoT Healthcare professionals Structural equation modeling UTAUT2

来源：评论

学校读者我要写书评

暂无评论

Framework for Effective Utilization of Distributed Scrum in software Projects

引用

computer Systems science & engineering 2023年第1期44卷 407-422页

作者： Basit Shahzad Wardah Naeem Awan Fazal-e-Amin Ahsanullah Abro Muhammad Shoaib Sultan Alyahya Department of Software Engineering National University of Modern LanguagesIslamabad44000Pakistan Department of Software Engineering College of Computer and Information SciencesKing Saud UniversityRiyadh11543Saudi Arabia Department of Computer Science Sukkur IBA UniversitySindh65200Pakistan Department of Information Systems College of Computer and Information SciencesKing Saud UniversityRiyadh11543Saudi Arabia

There is an emerging interest in using agile methodologies in Global software Development(GSD)to get the mutual benefits of both *** is currently admired by many development teams as an agile most known meth-odology and considered adequate for collocated *** the same time,stake-holders in GSD are dispersed by geographical,temporal,and socio-cultural *** to the controversial nature of Scrum and GSD,many significant challenges arise that might restrict the use of Scrum in *** conducted a Sys-tematic Literature Review(SLR)by following Kitchenham guidelines to identify the challenges that limit the use of Scrum in GSD and to explore the mitigation strategies adopted by practitioners to resolve the *** validate our reviewfindings,we conducted an industrial survey of 305 *** results of our study are consolidated into a research *** framework represents current best practices and recommendations to mitigate the identified distributed scrum challenges and is validated byfive experts of distributed *** of the expert review were found supportive,reflecting that the framework will help the stakeholders deliver sustainable products by effectively mitigating the identified challenges.

关键词： Global software development distributed Scrum sustainable development challenges mitigation strategies

来源：评论

学校读者我要写书评

暂无评论

A Systematic Review of Causal Analysis in Power Systems: Principles, Application, and Prospects 7

A Systematic Review of Causal Analysis in Power Systems: Pri...

引用

7th IEEE International Electrical and Energy Conference, CIEEC 2024

作者： Sun, Xiaorong Wang, Xuanheng Yan, Bincheng Wang, Baoqun Pan, Xueping Guo, Jinpeng School of Electrical and Power Engineering Hohai Univeristy Nanjing China College of Computer Science and Software Engineering Hohai University Nanjing China

ISBN: (纸本)9798350359558

Causal relationships are a scientific research method used to describe relationships between variable data, allowing for the excavation of the deep logic and operational mechanisms behind phenomena. The integration of causal relationships as a means of data analysis in the field of power systems improves the accuracy and efficiency of power systems analysis and operation. This paper presents the definition and origin of causal relationships and the principles of main causal analysis models. It overviews the application of causal relationships in power systems fault diagnosis, energy prediction, stability analysis, and the electricity market. Finally, it looks forward to the development prospects of causal relationships in combination with power physics models and artificial intelligence, promoting the construction of a reliable and intelligent new power systems under the penetration of renewable sources. © 2024 IEEE.

关键词： Systems analysis

来源：评论

学校读者我要写书评

暂无评论

An attention-based channel estimation network with residual mechanism for 5G OFDM systems 4

An attention-based channel estimation network with residual ...

引用

4th International Conference on Electronic Communication, computer science and Technology, ECCST 2024

作者： Jiang, Jinjing Huang, Weipeng Zhang, Xiao Yang, Kun Peng, Tong School of Information Engineering Zhejiang Ocean University China School of Computer Science and Software Engineering Shenzhen Institute of Information Technology China

Due to the significant increase in data transmission speed and gradual increase in Doppler frequency shift, channel estimation accuracy has become one of the most prioritized considerations in many cases. Specifically, we propose an attention-based channel estimation network with a residual mechanism, named TRCE-Net. In our design, the use of the multi-head attention mechanism at the input of the encoder is to obtain richer feature details. A linear series decoder module is implemented by leveraging multiple residual networks with skip connections. This novel design is capable of improving the channel estimation accuracy after deepening the network while effectively preventing degradation. To lessen the number of parameters that the network needs to learn, we introduce the bi-linear interpolation mechanism to replace the traditional up-sampling layer to mitigate both computational overhead and model complexity. According to the simulation results, our proposed TRCE-Net is superior to other state-of-the-art methods with regard to estimation accuracy and robustness to different maximum Doppler shifts, such as ResNet and InterpolateNet. © Published under licence by IOP Publishing Ltd.

关键词： Orthogonal frequency division multiplexing

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：