检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Han, Tingxu Huang, Shenghan Ding, Ziqi Sun, Weisong Feng, Yebo Fang, Chunrong Li, Jun Qian, Hanwei Wu, Cong Zhang, Quanjun Liu, Yang Chen, Zhenyu The State Key Laboratory for Novel Software Technology Nanjing University Jiangsu210093 China The School of Computer Science University of New South Wales NSW2052 Australia The School of Computer Science and Engineering Nanyang Technological University Nanyang639798 Singapore The Department of Computer Science The Director of the Network & Security Research Laboratory The University of Oregon United States

With the development of deep learning, self-supervised learning (SSL) has become particularly popular. SSL pre-trains an encoder (e.g., an image encoder) to learn generic features from a large amount of unlabeled data, releases it on a third-party platform, and allows any downstream task (e.g., an image classification task) to further fine-tune the encoder with task-specific data to derive the model for the task. However, despite the tremendous performance improvements it brings, SSL faces serious security risks. An adversary can poison pre-training datasets used to pre-train an encoder or publish a poisoned encoder directly to third-party platforms, which will also poison the model for any downstream task that uses these encoders. In this paper, we study a defense against poisoned encoders in SSL called distillation, which is a defense used in supervised learning originally. Distillation aims to distill knowledge from a given model (a.k.a the teacher net) and transfer it to another (a.k.a the student net). Now, we use it to distill benign knowledge from poisoned pre-trained encoders and transfer it to a new encoder, resulting in a clean pre-trained encoder. In particular, we conduct an empirical study on the effectiveness and performance of distillation against poisoned encoders. Using two state-of-the-art backdoor attacks against pre-trained image encoders and four commonly used image classification datasets, our experimental results show that distillation can reduce attack success rate from 80.87% to 27.51% while suffering a 6.35% loss in accuracy. Moreover, we investigate the impact of three core components of distillation on performance: Teacher net, student net, and distillation loss. By comparing 4 different teacher nets, 3 student nets, and 6 distillation losses, we find that fine-tuned teacher nets, warm-up-training-based student nets, and attention-based distillation loss perform best, respectively. Copyright © 2024, The Authors. All rights reserved.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

A Stochastic Proximal WMMSE for Ergodic Sum Rate Maximization

A Stochastic Proximal WMMSE for Ergodic Sum Rate Maximizatio...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Xiaotong Zhao Xi Wang Juncheng Wang Qingjiang Shi School of Software Engineering Tongji University Shanghai China Department of Computer Science Hong Kong Baptist University Hong Kong China Shenzhen Research Institute of Big Data Shenzhen China

We consider ergodic weighted sum rate (WSR) maximization in a massive multi-user multiple-input multiple-output system. Existing solutions iteratively minimize the average WSR based on all the historical information, and use bisection search to satisfy the power constraint at each iteration, resulting in both high storage burden and high computational complexity. In contrast, we propose an efficient stochastic proximal weighted minimum mean-square error (SPWMMSE) algorithm, which updates the precoder only based on the current single channel realization, without checking the power constraint at each iteration. Furthermore, we propose a novel proximal term to incorporate all the previous channel and surrogate function information in precoder updates. Our analysis shows that SPWMMSE converges to the stationary point of the original ergodic WSR maximization problem almost surely. Simulation results demonstrate the effectiveness of SPWMMSE over the current best alternatives.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Enhancing Diversity in Multi-Objective Feature Selection

Enhancing Diversity in Multi-Objective Feature Selection

引用

Congress on Evolutionary Computation

作者： Sevil Zanjani Miyandoab Shahryar Rahnamayan Azam Asilian Bidgoli Sevda Ebrahimi Masoud Makrehchi Nature-Inspired Computational Intelligence (NICI) Lab Department of Electrical Computer and Software Engineering Ontario Tech University Oshawa ON Canada Department of Engineering Brock University St. Catharines ON Canada Paculty of Science Wilfrid Laurier University Waterloo ON Canada

ISBN: (数字)9798350308365

ISBN: (纸本)9798350308372

Feature selection plays a pivotal role in the data preprocessing and model-building pipeline, significantly en-hancing model performance, interpretability, and resource efficiency across diverse domains. In population-based optimization methods, the generation of diverse individuals holds utmost importance for adequately exploring the problem landscape, particularly in highly multi-modal multi-objective optimization problems. Our study reveals that, in line with findings from sev-eral prior research papers, commonly employed crossover and mutation operations lack the capability to generate high-quality diverse individuals and tend to become confined to limited areas around various local optima. This paper introduces an augmen-tation to the diversity of the population in the well-established multi-objective scheme of the genetic algorithm, NSGA-II. This enhancement is achieved through two key components: the genuine initialization method and the substitution of the worst individuals with new randomly generated individuals as a re-initialization approach in each generation. The proposed multi-objective feature selection method undergoes testing on twelve real-world classification problems, with the number of features ranging from 2,400 to nearly 50,000. The results demonstrate that replacing the last front of the population with an equivalent number of new random individuals generated using the genuine initialization method and featuring a limited number of features substantially improves the population's quality and, consequently, enhances the performance of the multi-objective algorithm.

关键词： Sociology Pipelines Optimization methods Evolutionary computation Feature extraction Data models Distance measurement

来源：评论

学校读者我要写书评

暂无评论

Radiology Report Generation via Structured Knowledge-Enhanced Multi-modal Attention and Contrastive Learning

Radiology Report Generation via Structured Knowledge-Enhance...

引用

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Dexuan Xu Yanyuan Chen Jiayu Zhang Yiwei Lou Hanpin Wang Jing He Yu Huang National Engineering Research Center for S.E. School of Software & Micro. Peking University China School of Computer Science Peking University China Key Laboratory of High Confidece Software Technologies (MOE) School of Computer Science Peking University China Department of Rheumatology and Immunology Peking University People’s Hospital China National Engineering Research Center for Software Engineering Peking University China

The automated generation of radiology reports has attracted significant attention in the field of bioinformatics. Currently, the main limitations of this task include insufficient utilization of prior medical knowledge, lack of efficient knowledge fusion algorithms, and less distinctiveness between different generated reports. To address these issues, we propose a novel algorithm for radiology report generation, which includes Structured Knowledge-Enhanced Multi-modal Attention (SKEMA) and Dual-Branch Contrastive Learning (DBCL) for the first time. SKEMA aims to effectively bridge the gap between visual and prior knowledge by leveraging the high-order adjacency matrix of the knowledge graph to weightedly fuse image features and knowledge features. We enhance both features through masking, and use the original features and augmented features as positive and negative samples in the dual-branch contrastive learning (DBCL). DBCL increases the differences between positive and negative samples to avoid generating templated results, and enhances the robustness of the model. Finally, we conducted experiments to demonstrate the effectiveness of our model on two public radiology datasets, IU-Xray and MIMIC-CXR. Our model outperformed previous baseline methods on both datasets and achieved excellent evaluation scores.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Aligning Academic with Industrial Needs: Investigating DevOps 24

Aligning Academic with Industrial Needs: Investigating DevOp...

引用

Proceedings of the 28th International Conference on Evaluation and Assessment in software engineering

作者： Saima Rafi Muhammad Azeem Akbar Sajjad Mahmood School of Computing Engineering and the Built Environment Edinburgh Napier University Edinburgh United Kingdom Software Engineering LUT University Finland Information and Computer Science Department King Fahd University of Petroleum and Minerals Dhahran Saudi Arabia

ISBN: (纸本)9798400717017

Over recent years, DevOps has gained more attention within the software industry, owing to its pivotal role in facilitating continuous software delivery. The significant demand for DevOps practitioners necessitates substantial adjustments in conventional software engineering courses in higher educational institutions. Recent studies have highlighted DevOps principles, challenges, best practices, and tools to facilitate continuous delivery and deployment pipelines from an industrial perspective. As numerous universities are now introducing courses related to DevOps for students majoring in software engineering and computer science, this high demand for DevOps practitioners requires non-trivial adjustments in traditional software engineering courses and educational methodologies, including practices for teaching DevOps, training in DevOps, the level to which DevOps is currently taught and applied to measure the gap between teaching practices compared to industrial requirements. Our vision paper aims to highlight the crucial need for updated policies and tools to improve DevOps training in higher education, providing guidelines. By aligning academic curricula with industry standards, we can ensure students are better equipped for the demands of the workforce, fostering a seamless transition from academia to industry and promoting innovation in software engineering practices.

关键词： Challenges DevOps Guidelines Industry Practices Teaching

来源：评论

学校读者我要写书评

暂无评论

An Efficient Convolutional Multi-Scale Vision Transformer for Image Classification

An Efficient Convolutional Multi-Scale Vision Transformer fo...

引用

Image Processing, computer Vision and Machine Learning (ICICML), International Conference on

作者： Ji Zhang Zhihao Chen Yiyuan Ge Mingxin Yu School of Instrument Science and Opto-Electronics Engineering Beijing Information Science and Technology University Beijing China Computer School Beijing Information Science & Technology University Beijing China Department of Software Engineering Jilin University Changchun China

This paper introduces an innovative and efficient multi-scale Vision Transformer (ViT) for the task of image classification. The proposed model leverages the inherent power of transformer architecture and combines it with the concept of multi-scale processing generally used in convolutional neural networks (CNNs). The work aims to address the limitations of conventional ViTs which typically operate at a single scale, hence overlooking the hierarchical structure in visual data. The multi-scale ViT enhances classification performance by processing image features at different scales, effectively capturing both low-level and high-level semantic information. Extensive experimental results demonstrate the superior performance of the proposed model over standard ViTs and other state-of-the-art image classification methods, signifying the effectiveness of the multi-scale approach. This research opens new avenues for incorporating scale-variance in transformer-based models for improved performance in vision tasks.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Hybrid Makes Better: Hybrid Differential Privacy Medical Image Classification Based on Federated Learning

引用

IEEE Transactions on Dependable and Secure Computing 2025年

作者： Jiang, Shunrong Liu, Yingjie Wang, Zhi Li, Yonggang Chi, Haotian Du, Xiaojiang China University of Mining and Technology School of Computer Science Xuzhou221116 China Shanxi University School of Automation and Software Engineering Taiyuan030006 China Stevens Institute of Technology Department of Electrical and Computer Engineering HobokenNJ United States

Machine learning has the potential to revolutionize medical image classification. However, machine learning requires large medical datasets to improve accuracy, which will compromise patient privacy. Federated learning is a promising technique that protects patient privacy and improves the accuracy of medical image classification. Unfortunately, current research shows that federated learning faces the risk of privacy leakage. In this paper, we propose a privacy-preserving federal learning scheme via hybrid differential privacy for medical image classification (FHDM). Specifically, we construct a local hybrid differential privacy algorithm (LHDP) against patients' privacy leakage. This hybrid algorithm combines Gaussian and Laplace differential privacy without enlarging the privacy budget. We prove that the algorithm applies to the model parameter. Moreover, we design loss optimization and global optimization strategies on the algorithm to achieve higher accuracy in medical image classification. Finally, we validate FHDM in terms of privacy-preserving and model accuracy on real datasets. Experiments show that FHDM effectively protects privacy and improves the average accuracy by 9.60% compared to previous differential privacy schemes with the same medical image dataset and privacy budget. © 2004-2012 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Intelligent Gen: An Efficient On-Demand Access Protocol in MANETs using Network Simulator - 2

Intelligent Gen: An Efficient On-Demand Access Protocol in M...

引用

International Conference on Computing for Sustainable Global Development (INDIACom)

作者： T. Ganesan N. Murali S. Palani Murugan M. Chandrakumar Peter P. Anand Raj S. Manikandan Department of Computer Science & Engineering E.G.S. Pillay Engineering College Nagapattinam Tamil Nadu India Department of Artificial Intelligence & Data Science E.G.S. Pillay Engineering College Nagapattinam Tamil Nadu India Department of Software Engineering Periyar Maniammai Institute of Science & Technology Thanjavur Tamil Nadu India Department of Computer Science and Business Systems E.G.S. Pillay Engineering College Nagapattinam Tamil Nadu India Department of Information Technology E.G.S. Pillay Engineering College Nagapattinam Tamil Nadu India

Mobile Ad-Hoc network is a self-configuring and multi-hop collection of mobile devices for making successful communication. The wireless and mobile communication process are encountered based on bandwidth, process control and energy conception. Energy conception is considered and dynamically changed to rely on life time of batteries and power. The life time of battery get exhausted very quickly the nodes are moved and changed rapidly across the network. This paper noticed the problem of life time of power, batteries and energy consumption problem in various routing stages. The proposed Intelligent Gen Algorithm is used to get optimal solution for ensuring high robustness, flexibility and Energy consumption factor. It presents a common framework to balance various routing requirements in MANETs dynamically. The following are the main contributions includes, i. introducing Intelligent GEN Algorithms to classify the throughout bandwidth, delay, data ratio. ii. Various routing performance are evaluated and viewed by speed, packetize and simulation time. The comparison evaluated by select routing algorithm using NS-2. The simulation results are recorded and the performance is monitored by using on demand access protocl. The result clearly demonstrated the proposed Intelligent Gen algorithm used in self-organized manner by wireless channel.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Optimizing Performance of Cloud Computing Management Algorithm for High-Traffic Networks

Optimizing Performance of Cloud Computing Management Algorit...

引用

Disruptive Technologies (ICDT), International Conference on

作者： Venkata Ashok K Gorantla Shiva Kumar Sriramulugari Bhargavi Gorantla N. Yuvaraj Kiran Singh Solution Architect USA Sr. Lead Software Engineer USA Department of CSE Sri Shakthi Institute of Engineering and Technology Coimbatore INDIA Department of Computer Science and Engineering G.L. Bajaj Institute of Management Greater Noida INDIA

Cloud computing has revolutionized the way people use technology, enabling a variety of virtual services and applications in the corporate world. The improved cost-efficiency and flexibility of cloud computing have made it an attractive option for companies that require high-traffic networks. However, this has also increased the need for optimizing network performance in order to ensure reliable service and a pleasing experience for users. To this end, various algorithms have been proposed for cloud computing management. However, these algorithms often conflict with each other, leading to a decrease in performance. Thus, there is a need for a unified approach that can take into account different methods in a holistic manner. The optimization algorithms developed for cloud computing management are targeted towards improving resource utilization, failure handling, and load balancing. By properly allocating and balancing resources, the overhead associated with these tasks is greatly reduced. Furthermore, the intelligent use of techniques such as fault tolerance, capacity planning, and quality of service guarantees ensures that the users' experience within the cloud is of optimal quality. Parametric results from research on Cloud Computing management algorithm performance showed that, when correctly implemented, the algorithms can provide significant improvements to the management and performance of high-traffic networks. Specifically, performance improvements up to 40% were observed in resource allocation speed and energy consumption.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Music Genre Classification Using Convolutional Neural Network Method

Music Genre Classification Using Convolutional Neural Networ...

引用

International Conference on Information Technology Systems and Innovation (ICITSI)

作者： Intan Nurma Yulita Muhammad Ariq Farhansyah Mutyara Ino Suryana Firman Ardiansyah Muhammad Rasyid Ramdhani Mokhamad Arfan Wicaksono Department of Computer Science Universitas Padjadjaran Sumedang Indonesia Applied Bachelor in Digital Business Akademi Digital Bandung Bandung Indonesia Department of Electrical Engineering Universitas Padjadjaran Sumedang Indonesia Bachelor of Software Engineering Akademi Digital Bandung Bandung Indonesia

ISBN: (数字)9798331511470

ISBN: (纸本)9798331511487

Music genre classification is a prominent challenge in music information retrieval research. Most techniques for music genre classification employ machine learning algorithms to classify feature vectors extracted from sound recording segments. Several datasets have been utilized in experiments to evaluate classification accuracy in this domain. However, the datasets commonly used are relatively small, and much of the research on music genre classification continues to focus on these limited datasets. This research utilizes data from three main sources: (1) metadata derived from the Million Song Dataset, (2) the CD2C tagtraum genre annotations dataset as ground truth for genre labels, and (3) the Million Song Dataset mapping archive, which provides metadata mapping to SpotifyAPI, along with (4) acoustic features extracted via SpotifyAPI. The resulting dataset comprises 64,088 music tracks spanning five popular genres: Rock, Electronic, Jazz, Pop, and Rap. The objective of this research is to develop a model capable of predicting music genres based on acoustic features extracted through SpotifyAPI. To achieve this, a Convolutional Neural Network (CNN) architecture was implemented, with a focus on hyperparameter optimization, particularly in terms of the number of layers and the type of weight initialization. The optimal hyperparameters were identified as a two-layer CNN architecture utilizing Xavier Normal Initialization for weight initialization. The resulting model achieved a weighted F1-score of 0.7030 during testing.

关键词： Technological innovation Predictive models Feature extraction Rocks Acoustics Vectors Recording Convolutional neural networks Music information retrieval Testing

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：