检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Wu, Chen Zhang, Ruqing Guo, Jiafeng Chen, Wei Fan, Yixing de Rijke, Maarten Cheng, Xueqi CAS Key Lab of Network Data Science and Technology ICT CAS University of Chinese Academy of Sciences Beijing China University of Amsterdam Amsterdam Netherlands

Neural ranking models (NRMs) have achieved promising results in information retrieval. NRMs have also been shown to be vulnerable to adversarial examples. A typical Word Substitution Ranking Attack (WSRA) against NRMs was proposed recently, in which an attacker promotes a target document in rankings by adding human-imperceptible perturbations to its text. This raises concerns when deploying NRMs in real-world applications. Therefore, it is important to develop techniques that defend against such attacks for NRMs. In empirical defenses adversarial examples are found during training and used to augment the training set. However, such methods offer no theoretical guarantee on the models’ robustness and may eventually be broken by other sophisticated WSRAs. To escape this arms race, rigorous and provable certified defense methods for NRMs are needed. To this end, we first define the Certified Top-K Robustness for ranking models since users mainly care about the top ranked results in real-world scenarios. A ranking model is said to be Certified TopK Robust on a ranked list when it is guaranteed to keep documents that are out of the top K away from the top K under any attack. Then, we introduce a Certified Defense method, named CertDR, to achieve certified top-K robustness against WSRA, based on the idea of randomized smoothing. Specifically, we first construct a smoothed ranker by applying random word substitutions on the documents, and then leverage the ranking property jointly with the statistical property of the ensemble to provably certify top-K robustness. Extensive experiments on two representative web search datasets demonstrate that CertDR can significantly outperform state-of-the-art empirical defense methods for ranking models. © 2022, CC0.

关键词： network security

来源：评论

学校读者我要写书评

暂无评论

Class Incremental Learning for Knowledge Acquisition Based on Prototype network Gating Mechanism

Class Incremental Learning for Knowledge Acquisition Based o...

引用

Medical Artificial Intelligence (MedAI), IEEE International Conference on

作者： Yuming Feng Xiaolong Jin Haozhe Liang Jiafeng Guo School of Computer Science and Technology University of Chinese Academy of Sciences CAS Key Laboratory of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences Institute of Systems Engineering Academy of Military Science

ISBN: (数字)9798350377613

ISBN: (纸本)9798350377620

With the development of information technology, the medical field has accumulated a large amount of medical data, which continues to grow. Efficient extraction and effective utilization of knowledge from this vast amount of medical data are crucial for realizing its immense value. Traditional knowledge acquisition algorithms face challenges in effectively identifying and acquiring new types of knowledge in practical applications. To address this, this paper proposes a class-incremental learning method based on a prototype network gating mechanism, called PGMoE. This method integrates multiple experts through a prototype network-based gating mechanism, dynamically adjusting the model's response to new and old knowledge, thereby effectively mitigating the issue of catastrophic forgetting in class-incremental learning. PGMoE not only improves the stability and accuracy of the model in continuous learning processes but also ensures the model's ability to remember historical knowledge. Comparative experiments on multiple datasets with existing class-incremental learning methods demonstrate that PGMoE has significant performance advantages in handling continuous learning tasks.

关键词： Learning systems Accuracy Incremental learning Knowledge acquisition Scalability Heuristic algorithms Prototypes Logic gates Stability analysis Information technology

来源：评论

学校读者我要写书评

暂无评论

Multi-chain Storage Scheme Based on Hash Calendar and Recursive Zero-knowledge Proof 2

Multi-chain Storage Scheme Based on Hash Calendar and Recurs...

引用

2nd International Conference on Computer science and Blockchain, CCSB 2022

作者： Wang, Chao Yao, Zhongyuan Zhang, Yong Guo, Shangkun Zhao, Lindong Si, Xueming Henan International Joint Laboratory of Blockchain and Data Sharing Zhengzhou Henan China Henan Key Laboratory of Network Cryptography Technology Zhengzhou Henan China Zhongyuan University of Technology The Frontier Information Technology Research Institute China Shanghai Jiao Tong University Department of Computer Science and Engineer Shanghai China

ISBN: (纸本)9798350331882

Blockchain technology is currently a research hotspot in the computer field. However, with the rapid development of blockchain technology, poor storage scalability limits its further application. In response to this problem, this paper proposes a multi-chain storage scheme based on hash calendar and recursive zero-knowledge proof, which consists of two parts: off-chain and on-chain. The off-chain part uses hash calendar technology to store account information, and through recursive zero-knowledge proof of account state transitions, ensures the verifiability of off-chain state transitions and improves the security of the solution;and sets up proof generators and transactions The two roles of the executor improve the efficiency of transaction execution. A dynamically addable main and sub-chain array is designed on the chain. The sub-chains store the brief information of transactions and the world state. The main chain is responsible for verifying the correctness of off-chain state transitions. On-chain and off-chain collaborative storage reduces the space occupied by data., to realize the storage expansion of the blockchain. Finally, simulation experiments show that this scheme can effectively realize the storage expansion of blockchain. © 2022 IEEE.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Thematic Event Extraction Based on Event-Related Sentence Detection

Thematic Event Extraction Based on Event-Related Sentence De...

引用

Medical Artificial Intelligence (MedAI), IEEE International Conference on

作者： Xing Zhang Yucan Guo Xiaolong Jin Haozhe Liang Saiping Guan School of Computer Science and Technology University of Chinese Academy of Sciences CAS Key Laboratory of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences Institute of Systems Engineering Academy of Military Science

Existing works in event extraction typically extract event arguments within the sentence scope. However, besides the sentence level, events may also be naturally presented at the document level. A document-level event usually reflects, to some extent, the theme (i.e., the main content) of the document (e.g., electronic medical records and news articles), which is thus referred to as the thematic event. Thematic Event Extraction (TEE) aims to extract the arguments of thematic events. TEE faces a major challenge, i.e., the sparsity and dispersion of arguments, which means that the arguments of a thematic event are dispersed in different sentences of the document. To overcome this challenge, we propose an Event-related Sentence Detection based TEE model, called ESDTEE, which first detects the sentences related to the thematic event and then extracts the arguments only within these detected sentences using existing models. Extensive experiments with comprehensive analyses demonstrate the effectiveness of ESDTEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-View Graph Embedding Learning for Image Co-Segmentation and Co-Localization

引用

IEEE Transactions on Circuits and Systems for Video technology 2024年第6期34卷 4942-4956页

作者： Huang, Aiping Li, Lijian Zhang, Le Niu, Yuzhen Zhao, Tiesong Lin, Chia-Wen Fuzhou University Fujian Key Laboratory for Intelligent Processing and Wireless Transmission of Media Information College of Physics and Information Engineering Fuzhou350108 China Fuzhou University Fujian Key Laboratory of Network Computing and Intelligent Information Processing College of Computer and Data Science Fuzhou350108 China University of Electronic Science and Technology of China School of Information and Communication Engineering Chengdu611731 China Fuzhou University Fujian Key Laboratory for Intelligent Processing and Wireless Transmission of Media Information College of Physics and Information Engineering The Fujian Science and Technology Innovation Laboratory for Optoelectronic Information Fuzhou350108 China Institute of Communications Engineering National Tsing Hua University Department of Electrical Engineering Hsinchu30013 Taiwan

Image co-segmentation and co-localization exploit inter-image information to identify and extract foreground objects with a batch mode. However, they remain challenging when confronted with large object variations or complex backgrounds. This paper proposes a multi-view graph embedding (MV-Gem) learning scheme which integrates diversity, robustness and discernibility of object features to alleviate this phenomenon. To encourage the diversity, the deep co-information containing both low-layer general representations and high-layer semantic information is generated to form a multi-view feature pool for comprehensive co-object description. To enhance the robustness, a multi-view adaptive weighted learning is formulated to fuse the deep co-information for feature complementation. To ensure the discernibility, the graph embedding and sparse constraint are embedded into the fusion formulation for feature selection. The former aims to inherit important structures from multiple views, and the latter further selects important features to restrain irrelevant backgrounds. With these techniques, MV-Gem gradually recovers all co-objects through optimization iterations. Extensive experimental results on real-world datasets demonstrate that MV-Gem is capable of locating and delineating co-objects in an image group. © 1991-2012 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

LSMM: A Life-cycle Security Management Mechanism for Secure Image Sharing 2

LSMM: A Life-cycle Security Management Mechanism for Secure ...

引用

2nd International Conference on Computer science and Blockchain, CCSB 2022

作者： Zhang, Yaoyao Yao, Zhongyuan Pan, Heng Si, Xueming Henan International Joint Laboratory of Blockchain and Data Sharing Zhengzhou Henan China Henan Key Laboratory of Network Cryptography Technology Zhengzhou Henan China The Frontier Information Technology Research Institute Zhongyuan University of Technology China Shanghai Jiao Tong University Department of Computer Science and Engineer Shanghai China

ISBN: (纸本)9798350331882

During the storing, transferring, processing, and exchanging phases of the full life-cycle of secure image sharing, there may exist security issues such as information leakage, infringement and abuse. To address those problems, a life-cycle security management mechanism is proposed. By utilizing the Commutative Encryption and Watermarking (CEW), the mechanism ensures that the ownership and confidentiality of one image can hardly be tampered with and compromised. Besides, with the help of the blockchain, the mechanism is enabled to record any changes during the whole process of the image sharing, which makes copyright traceability possible and thus reduces the occurrence of infringement and abuse. The experiment result shows that, in our mechanism, the quality of plaintext image with watermark obtained by decryption is high, and the invisibility of watermark is preserved. Also, the efficient analysis and experimenting of the mechanism exhibits that it is feasible in practice. © 2022 IEEE.

关键词： Life cycle

来源：评论

学校读者我要写书评

暂无评论

Attribute Propagation Enhanced Community Detection Model for Bitcoin De-anonymizing 3rd

Attribute Propagation Enhanced Community Detection Model for...

引用

3rd International Conference on Machine Learning for Cyber Security, ML4CS 2020

作者： Wang, Jiming Xie, Xueshuo Fang, Yaozheng Lu, Ye Li, Tao Wang, Guiling College of Computer Science Nankai University Tianjin300350 China College of Cyber Science Nankai University Tianjin300350 China Tianjin Key Laboratory of Network and Data Security Technology Tianjin300350 China New Jersey Institute of Technology NewarkNJ07102 United States

ISBN: (纸本)9783030622220

Bitcoin is a kind of decentralized cryptocurrency on a peer-to-peer network. Anonymity makes Bitcoin widely used in online payment but it is a disadvantage for regulatory purposes. We aim to de-anonymize Bitcoin to assist regulation. Many previous studies have used heuristic clustering or machine learning to analyze historical transactions and identify user behaviors. However, the accuracy of user identification is not ideal. Heuristic clustering only uses the topological structure of the transaction graph and ignores many transaction information, and supervised machine learning methods are limited by the size of labeled datasets. To identify user behaviors, we propose a community detection model based on attribute propagation, combining the topological structure of the transaction graph and additional transaction information. We first parse the transaction data of public ledger and construct a bipartite graph to describe correlations between addresses and transactions. We also extract address attributes from historical transactions to construct an attributed graph with the previous bipartite graph. Then, we design an adaptive weighted attribute propagation algorithm named AWAP running on the attributed graph to classify bitcoin addresses, and further identify user behaviors. Extensive experiments highlight that the proposed detection model based on AWAP achieves 5% higher accuracy on average compared to state-of-the-art address classification methods in Bitcoin. AWAP also achieves 25% higher F-score on average compared to previous community detection algorithms on two datasets. © 2020, Springer Nature Switzerland AG.

关键词： Bitcoin

来源：评论

学校读者我要写书评

暂无评论

Augmentation-Aware Self-Supervision for data-Efficient GAN Training

arXiv

引用

arXiv 2022年

作者： Hou, Liang Cao, Qi Yuan, Yige Zhao, Songtao Ma, Chongyang Pan, Siyuan Wan, Pengfei Wang, Zhongyuan Shen, Huawei Cheng, Xueqi CAS Key Laboratory of AI Safety and Security Institute of Computing Technology Chinese Academy of Sciences China CAS Key Laboratory of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences China University of Chinese Academy of Sciences China Kuaishou Technology China

Training generative adversarial networks (GANs) with limited data is challenging because the discriminator is prone to overfitting. Previously proposed differentiable augmentation demonstrates improved data efficiency of training GANs. However, the augmentation implicitly introduces undesired invariance to augmentation for the discriminator since it ignores the change of semantics in the label space caused by data transformation, which may limit the representation learning ability of the discriminator and ultimately affect the generative modeling performance of the generator. To mitigate the negative impact of invariance while inheriting the benefits of data augmentation, we propose a novel augmentation-aware self-supervised discriminator that predicts the augmentation parameter of the augmented data. Particularly, the prediction targets of real data and generated data are required to be distinguished since they are different during training. We further encourage the generator to adversarially learn from the self-supervised discriminator by generating augmentation-predictable real and not fake data. This formulation connects the learning objective of the generator and the arithmetic − harmonic mean divergence under certain assumptions. We compare our method with state-of-the-art (SOTA) methods using the class-conditional BigGAN and unconditional StyleGAN2 architectures on data-limited CIFAR-10, CIFAR-100, FFHQ, LSUN-Cat, and five low-shot datasets. Experimental results demonstrate significant improvements of our method over SOTA methods in training data-efficient GANs. Copyright © 2022, The Authors. All rights reserved.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Querying a matrix through matrix-vector products 46

Querying a matrix through matrix-vector products

引用

46th International Colloquium on Automata, Languages, and Programming, ICALP 2019

作者： Sun, Xiaoming Woodruff, David P. Yang, Guang Zhang, Jialin CAS Key Lab of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China Carnegie Mellon University PittsburghPA United States Institute of Computing Technology Chinese Academy of Sciences Beijing China Conflux Beijing China

ISBN: (纸本)9783959771092

We consider algorithms with access to an unknown matrix M ∈ Fn×d via matrix-vector products, namely, the algorithm chooses vectors v1,..., vq, and observes Mv1,..., Mvq. Here the vi can be randomized as well as chosen adaptively as a function of Mv1,..., Mvi−1. Motivated by applications of sketching in distributed computation, linear algebra, and streaming models, as well as connections to areas such as communication complexity and property testing, we initiate the study of the number q of queries needed to solve various fundamental problems. We study problems in three broad categories, including linear algebra, statistics problems, and graph problems. For example, we consider the number of queries required to approximate the rank, trace, maximum eigenvalue, and norms of a matrix M;to compute the AND/OR/Parity of each column or row of M, to decide whether there are identical columns or rows in M or whether M is symmetric, diagonal, or unitary;or to compute whether a graph defined by M is connected or triangle-free. We also show separations for algorithms that are allowed to obtain matrix-vector products only by querying vectors on the right, versus algorithms that can query vectors on both the left and the right. We also show separations depending on the underlying field the matrix-vector product occurs in. For graph problems, we show separations depending on the form of the matrix (bipartite adjacency versus signed edge-vertex incidence matrix) to represent the graph. Surprisingly, this fundamental model does not appear to have been studied on its own, and we believe a thorough investigation of problems in this model would be beneficial to a number of different application areas. © Xiaoming Sun, David P. Woodruff, Guang Yang, and Jialin Zhang;licensed under Creative Commons License CC-BY

关键词： Matrix algebra

来源：评论

学校读者我要写书评

暂无评论

Incentive Scheme for Shared Parking Space based on NFT 2

Incentive Scheme for Shared Parking Space based on NFT

引用

2nd International Conference on Computer science and Blockchain, CCSB 2022

作者： Cai, Fugang Pan, Heng Si, Xueming Wu, Zhili Qian, Haiyang Henan International Joint Laboratory of Blockchain and Data Sharing Zhengzhou Henan China Henan Key Laboratory of Network Cryptography Technology Zhengzhou Henan China Zhongyuan University of Technology The Frontier Information Technology Research Institute China Shanghai Jiao Tong University Department of Computer Science and Engineer Shanghai China

ISBN: (纸本)9798350331882

The lack of centralized management causes problems in the system of distributed parking spaces, such as unsecured revenues, poor service quality and malicious valuation. Therefore, this paper proposes a parking sharing scheme based on non-fungible tokens. Firstly, we propose the calculation method of parking space renter reputation value, parking space provider reputation value and the reputation incentive algorithm of parking space sharing, which can ensure the service quality of distributed parking space share and prevent malicious evaluation and implement bidirectional constraints on parking space provider and renter. Secondly, we propose a parking space sharing model with non-fungible tokens as sharing credentials to map parking space assets to on-chain assets via non-fungible tokens and introduce timers based on hash time locks. We propose an intelligent parking rental contract based on improved hash lock to guarantee the revenue of parking space providers in a distributed environment. Finally, experiments based on Ethereum show that the scheme can realize shared parking space at a lower cost;the proposed incentive algorithm can effectively prevent parking space providers and renters from misbehaving. © 2022 IEEE.

关键词： Smart contract

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：