检索结果-内蒙古大学图书馆

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Zaixi Zhang Qi Liu Zhicai Wang Zepu Lu Qingyong Hu Anhui Province Key Lab of Big Data Analysis and Application School of Computer Science and Technology University of Science and Technology of China State Key Laboratory of Cognitive Intelligence Hefei Anhui China University of Science and Technology of China Hong Kong University of Science and Technology

Deep neural networks (DNNs) are recently shown to be vulnerable to backdoor attacks, where attackers embed hidden backdoors in the DNN model by injecting a few poisoned examples into the training dataset. While extensive efforts have been made to detect and remove backdoors from backdoored DNNs, it is still not clear whether a backdoor-free clean model can be directly obtained from poisoned datasets. In this paper, we first construct a causal graph to model the generation process of poisoned data and find that the backdoor attack acts as the confounder, which brings spurious associations between the input images and target labels, making the model predictions less reliable. Inspired by the causal understanding, we propose the Causality-inspired Backdoor Defense (CBD), to learn deconfounded representations for reliable classification. Specifically, a backdoored model is intentionally trained to capture the confounding effects. The other clean model dedicates to capturing the desired causal effects by minimizing the mutual information with the confounding representations from the backdoored model and employing a sample-wise re-weighting scheme. Extensive experiments on multiple benchmark datasets against 6 state-of-the-art attacks verify that our proposed defense method is effective in reducing backdoor threats while maintaining high accuracy in predicting benign samples. Further analysis shows that CBD can also resist potential adaptive attacks. The code is available at https://***/zaixizhang/CBD.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Network configuration approach based on the reliability service

Network configuration approach based on the reliability serv...

引用

IEEE International Symposium on Information (IT) in Medicine and Education, ITME

作者： Kuang Fengfei Lin Yaojm Yu Xiehua Chen Meilong Key Laboratory of Big Data and Artificial Intelligence in University of Fujian Province MinNan Science and Technology University Fujian Quanzhou China School of Computer Science Lab of Data Science and Intelligence Application Minnan Normal University Zhangzhou China Key Laboratory of Big Data and Artificial Intelligence University of Fujian Province MinNan Science and Technology University Quanzhou China

Wireless communication network plays an important role in various industries. One of the key concern in wireless communication is network security. This paper proposes an optimal transmission lines assignment with maximal reliabilities in the wireless communication. Firstly, a reliability evaluation metric with a transmission line assignment is proposed for calculating the reliability service under some constraints. Secondly, based on the reliability, non-dominated sorting Genetic algorithm is used for working out the optimal solution via considering the network security. Some simulated experiments are conducted that the proposed approach is able to obtain the reasonable solution under the time constraints.

关键词： Wireless communication Power transmission lines Computer network reliability Computational modeling Network security Reliability engineering Time factors

来源：评论

学校读者我要写书评

暂无评论

Blockchain-based data Sharing Framework for Malaysia Government Aid Management System

Blockchain-based Data Sharing Framework for Malaysia Governm...

引用

2023 International Conference on Digital applications, Transformation and Economy, ICDATE 2023

作者： Azmi, Azman Yahya, Farashazillah Moung, Ervin Gubin Sallehudin, Hasimi Utomo, Rio Guntur Azman, Nur Afrina Universiti Malaysia Sabah Advanced Machine Intelligence Research Lab Faculty of Computing and Informatics Sabah Kota Kinabalu Malaysia Universiti Malaysia Sabah Data Technologies and Application Research Lab Faculty of Computing and Informatics Sabah Kota Kinabalu Malaysia Universiti Kebangsaan Malaysia Centre for Software Technology and Management Faculty of Information Science and Technology Malaysia Telkom University School of Computing Bandung Indonesia Foundation Study and Language Institute University of Reading Malaysia Johor Iskandar Puteri Malaysia

ISBN: (纸本)9798350310689

The Covid-19 pandemic has impacted the world economy, and Malaysia is no exception. Due to the closure of businesses, the rate of unemployment has been recorded at an all-time high. The Malaysian Government through its agencies is assisting citizens through an online platform for incentives and an aid-giving platform. The current architecture of data sharing is using centralized, organization-oriented authentication and is not immutable for audit capability. In addition, the blockchain network is still a proof of concept in the digital government ecosystem. This paper studies the current aid data-sharing ecosystem and analyzes the potential of having a state government consortium blockchain framework that will be used in the Government Network Infrastructure for decentralized, transparent data sharing with audit capability. Several works involving Blockchain adoption based on domain, Blockchain consortium architecture and Government applications using blockchain are being presented. The need for a framework in implementing decentralized data sharing using consortium blockchain architecture for multi-agencies data sharing focusing on Malaysia Government Aid Management is discussed in this paper to provide the opportunity for the research domain. The findings in this paper will be used for future work on the requirement of a proposed government multi-layer consortium blockchain framework for data sharing in the Malaysian Government specifically for the Johore State Government ecosystem. © 2023 IEEE.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Reversible data Hiding Using Convolutional Neural Network and Digital Signal Processing Techniques

Reversible Data Hiding Using Convolutional Neural Network an...

引用

IEEE International Symposium on Information (IT) in Medicine and Education, ITME

作者： Peng Qingmei Li Shaozi Lin Yaojin Yu Xiehua Key Laboratory of Big Data and Artificial Intelligence University of Fujian Province MinNan Science and Technology University Quanzhou China School of Informatics Xiamen University Xiamen China School of Computer Science Lab of Data Science and Intelligence Application Minnan Normal University Zhangzhou China

Reversible data hiding is a data hiding process in which data is encrypted into an image as a means of authentication, and hidden data is used to hide the original image, and RDH is often used in areas where data authenticity and data preservation are important, such as legal/criminal investigations, medical images. In this study, this paper designed a novel reversible data hiding system that utilizes the Steghide algorithm to encrypt images and hide data, as well as an artificial neural network to recover embedded text and structure. Based on the analysis of the test results, it was concluded that the RDH scheme proposed in this article has the same or higher system accuracy and reliability as the existing RDH scheme. To further refine the system results in future studies, this paper proposes to increase the number of training iterations and the number of images in the training dataset to further improve the accuracy and reliability of the development.

关键词： Training System testing Signal processing algorithms Digital signal processing Reliability Cryptography Time factors

来源：评论

学校读者我要写书评

暂无评论

Sustainable Self-evolution Adversarial Training

arXiv

引用

arXiv 2024年

作者： Wang, Wenxuan Wang, Chenglei Qi, Huihui Ye, Menghao Qian, Xuelin Wang, Peng Zhang, Yanning School of Computer Science Northwestern Polytechnical University National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology Shaanxi Xi’an China School of Automation Northwestern Polytechnical University The BRain and Artificial INtelligence Lab Shaanxi Xi’an China

With the wide application of deep neural network models in various computer vision tasks, there has been a proliferation of adversarial example generation strategies aimed at deeply exploring model security. However, existing adversarial training defense models, which rely on single or limited types of attacks under a one-time learning process, struggle to adapt to the dynamic and evolving nature of attack methods. Therefore, to achieve defense performance improvements for models in long-term applications, we propose a novel Sustainable Self-Evolution Adversarial Training (SSEAT) framework. Specifically, we introduce a continual adversarial defense pipeline to realize learning from various kinds of adversarial examples across multiple stages. Additionally, to address the issue of model catastrophic forgetting caused by continual learning from ongoing novel attacks, we propose an adversarial data replay module to better select more diverse and key relearning data. Furthermore, we design a consistency regularization strategy to encourage current defense models to learn more from previously trained ones, guiding them to retain more past knowledge and maintain accuracy on clean samples. Extensive experiments have been conducted to verify the efficacy of the proposed SSEAT defense method, which demonstrates superior defense performance and classification accuracy compared to competitors. Copyright © 2024, The Authors. All rights reserved.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Medical Image Fusion Solid Works Supporting Multiple Techniques with Feature-Level Transforms 2

Medical Image Fusion Solid Works Supporting Multiple Techniq...

引用

2nd International Conference on Sustainable Computing and data Communication Systems, ICSCDS 2023

作者： Mutha, Rakhi Pathak, Devendra Nath Ahuja, Vandana Bhandari, Prabhakar Ramesh Rahi, Pankaj Mamodiya, Udit Lovely Professional University Department of Computer Application Punjab India Poornima College of Engineering Department of Computer Science and Engineering Rajasthan Jaipur India Swami Vivekanand Institute of Engineering and Technology Department of Computer Science and Engineering Punjab India Sevadal Mahila Mahavidyalaya Department of Microbiology Umred Road Maharashtra Nagpur India Poornima Institute of Engineering and Technology Department of Artificial Intelligence & Data Science Rajasthan Jaipur India Piet Aicte Idea Lab Poornima Institute of Engineering and Technology Department of Electrical Engineering Rajasthan Jaipur India

ISBN: (纸本)9781665491990

The use of group classifiers on similar features has resulted in computationally expensive and tedious, making them unsuitable for online applications. The clustering techniques are evaluated for both inter-subject and intra-subject variations, and it is discovered that intra-subject order execution is superior. Another approach to dealing with object recognition from medical images is considered. This is known as similarity matching, and it is a computationally simple process as compared to costly and complex characterization computations. The major challenges that occur in the image processing are high noise levels, limited image resolution, and geometric deformations throughout the image. © 2023 IEEE.

关键词： Image resolution

来源：评论

学校读者我要写书评

暂无评论

Full-atom protein pocket design via iterative refinement 23

Full-atom protein pocket design via iterative refinement

引用

Proceedings of the 37th International Conference on Neural Information Processing Systems

作者： Zaixi Zhang Zepu Lu Zhongkai Hao Marinka Zitnik Qi Liu Anhui Province Key Lab of Big Data Analysis and Application School of Computer Science and Technology University of Science and Technology of China and State Key Laboratory of Cognitive Intelligence Hefei Anhui China and Harvard University Anhui Province Key Lab of Big Data Analysis and Application School of Computer Science and Technology University of Science and Technology of China and State Key Laboratory of Cognitive Intelligence Hefei Anhui China Dept. of Comp. Sci. and Tech. Institute for AI THBI Lab BNRist Center Tsinghua-Bosch Joint ML Center Tsinghua Harvard University

The design and optimization of functional proteins that bind specific ligand molecules is paramount in therapeutics and bio-engineering. A critical yet formidable task in this endeavor is the design of the protein pocket, which is the cavity region of the protein where the ligand binds. Current methods are plagued by inefficient generation, inadequate context modeling of the ligand molecule, and the inability to generate side-chain atoms. Here, we present the Full-Atom Iterative Refinement (FAIR) method, designed to address these challenges by facilitating the co-design of protein pocket sequences, specifically residue types, and their corresponding 3D structures. FAIR operates in two steps, proceeding in a coarse-to-fine manner (transitioning from protein backbone to atoms, including side chains) for a full-atom generation. In each iteration, all residue types and structures are simultaneously updated, a process termed full-shot refinement. In the initial stage, the residue types and backbone coordinates are refined using a hierarchical context encoder, complemented by two structure refinement modules that capture both inter-residue and pocket-ligand interactions. The subsequent stage delves deeper, modeling the side-chain atoms of the pockets and updating residue types to ensure sequence-structure congruence. Concurrently, the structure of the binding ligand is refined across iterations to accommodate its inherent flexibility. Comprehensive experiments show that FAIR surpasses existing methods in designing superior pocket sequences and structures, producing average improvement exceeding 10% in AAR and RMSD metrics.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Backdoor Defense via Deconfounded Representation Learning

arXiv

引用

arXiv 2023年

作者： Zhang, Zaixi Liu, Qi Wang, Zhicai Lu, Zepu Hu, Qingyong Anhui Province Key Lab of Big Data Analysis and Application School of Computer Science and Technology University of Science and Technology of China China State Key Laboratory of Cognitive Intelligence Anhui Hefei China Hong Kong University of Science and Technology Hong Kong University of Science and Technology of China China

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

A review of distributed statistical inference

引用

Statistical Theory and Related Fields 2022年第2期6卷 89-99页

作者： Yuan Gao Weidong Liu Hansheng Wang Xiaozhou Wang Yibo Yan Riquan Zhang School of Statistics and Key Laboratory of Advanced Theory and Application in Statistics and Data Science–MOE East China Normal UniversityShanghaiPeople’s Republic of China School of Mathematical Sciences and Key Lab of Articial Intelligence–MOE Shanghai Jiao Tong UniversityShanghaiPeople’s Republic of China Guanghua School of Management Peking UniversityBeijingPeople’s Republic of China

The rapid emergence of massive datasets in various fields poses a serious challenge to tra-ditional statistical ***,it provides opportunities for researchers to develop novel *** by the idea of divide-and-conquer,various distributed frameworks for statistical estimation and inference have been *** were developed to deal with large-scale statistical optimization *** paper aims to provide a comprehensive review for related *** includes parametric models,nonparametric models,and other frequently used *** key ideas and theoretical properties are *** trade-off between communication cost and estimate precision together with other concerns is discussed.

关键词： Distributed computing divide-and-conquer communication-efficiency shrinkage methods nonparametric estimation principal component analysis feature screening bootstrap

来源：评论

学校读者我要写书评

暂无评论

Teaching resource recommendation for OBE based on kernel canonical correlation analysis

Teaching resource recommendation for OBE based on kernel can...

引用

IEEE International Symposium on Information (IT) in Medicine and Education, ITME

作者： Wu Changqian Li Shaozi Lin Yaojin Yu Xiehua Key Laboratory of Big Data and Artificial Intelligence in University of Fujian Province MinNan Science and Technology University José Rizal University Quanzhou China School of Informatics Xiamen University Xiamen China School of Computer Science Lab of Data Science and Intelligence Application Minnan Normal University Zhangzhou China Key Laboratory of Big Data and Artificial Intelligence in University of Fujian Province MinNan Science and Technology University Quanzhou China

data mining is used to extract information from OBE teaching resources, then the teaching resources and user personality database are established to improve the effectiveness of resources recommendation on the online learning platform. The properties of teaching resources and user personality traits are analyzed using the kernel canonical correlation analysis algorithm, and teaching materials with a high correlation coefficient are chosen to recommend to users. The simulation results demonstrate that this proposed method has higher accuracy and more accurate recommended resources when compared to the widely used teaching resource recommendation techniques.

关键词： Correlation coefficient Correlation databases Simulation Education data mining Kernel

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：