检索结果-内蒙古大学图书馆

2022 Findings of the Association for Computational Linguistics: NAACL 2022

作者： Wang, Xin Wang, Yasheng Wan, Yao Wang, Jiawei Zhou, Pingyi Li, Li Wu, Hao Liu, Jin School of Computer Science Wuhan University China Huawei Noah's Ark Lab Hong Kong School of Computer Sci. & Tech. Huazhong University of Science and Technology China Faculty of Information Technology Monash University Australia School of Information Science and Engineering Yunnan University China

ISBN: (纸本)9781955917766

Recent years have witnessed increasing interest in code representation learning, which aims to represent the semantics of source code into distributed vectors. Currently, various works have been proposed to represent the complex semantics of source code from different views, including plain text, Abstract Syntax Tree (AST), and several kinds of code graphs (e.g., Control/Data Flow Graph). However, most of them only consider a single view of source code independently, ignoring the correspondences among different views. In this paper, we propose to integrate different views with the naturallanguage description of source code into a unified framework with Multi-View contrastive Pre-training, and name our model as CODEMVP. Specifically, we first extract multiple code views using compiler tools, and learn the complementary information among them under a contrastive learning framework. Inspired by the type checking in compilation, we also design a fine-grained type inference objective in the pre-training. Experiments on three downstream tasks over five datasets demonstrate the superiority of CODE-MVP when compared with several state-of-the-art baselines. For example, we achieve 2.4/2.3/1.1 gain in terms of MRR/MAP/Accuracy metrics on natural language code retrieval, code similarity, and code defect detection tasks, respectively. © Findings of the Association for Computational Linguistics: NAACL 2022 - Findings.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

FusTP-FL: Enhancing Differential Federated Learning through Personalized Layers and Data Transformation

FusTP-FL: Enhancing Differential Federated Learning through ...

引用

IEEE International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom)

作者： Xiong Yan Kedong Yan Chanying Huang Dan Yin Shan Xiao School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China School of Comp. Science & Engineering Nanjing Univ. of Science & Technology Nanjing China School of Elec. & Inf. Engineering Beijing Univ. of Civil Engineering & Architecture Beijing China Fiberhome Comm. Tech. Co. Wuhan China

ISBN: (数字)9798331506209

ISBN: (纸本)9798331506216

Federated Learning enables multiple clients to collaboratively train a model without sharing their individual data, thereby protecting local data privacy. However, attackers, such as untrusted servers, can still compromise the privacy of clients’ local training data through various inference attacks. One feasible approach to protect client privacy during training is the incorporation of differential privacy. Nevertheless, achieving an ideal level of privacy protection with differential privacy often degrades the model’s performance, significantly reducing its accuracy. To enhance model accuracy while minimizing the additional client heterogeneity introduced by differential privacy, this paper proposes a method that integrates personalized layers and data transformations, FusTP-FL. The core of our FusTP-FL is the incorporation of personalized layers and personalized data transformations within the client’s local training model, which further reduces client heterogeneity and improves model accuracy. We evaluated the model’s accuracy on six common datasets; experimental results demonstrate that the proposed FusTP-FL effectively enhances model accuracy across two different differential privacy modes (CDP and LDP), increasing it by up to 45%. Furthermore, we show that compared to PRIVATEFL, our method achieves lower client heterogeneity.

关键词： Training Differential privacy Privacy Accuracy Federated learning Computational modeling Training data Data models Servers Security

来源：评论

学校读者我要写书评

暂无评论

DLSCA: Improving Cross-Device Side-Channel Analysis Using Device Discrepancy Correction

DLSCA: Improving Cross-Device Side-Channel Analysis Using De...

引用

Smart World Congress (SWC), IEEE

作者： Fanliang Hu Yi Li Haowen Tan Shan Jiang Zhiyuan Xiao School of Computer Science Nanjing University of Information Science and Technology Nanjing China School of Information Science and Engineering Zhejiang Sci-Tech University Hangzhou China

In this paper, we focus on the problem of Transfer Learning (TL) in Side-Channel Analysis (SCA) due to differences between the source device and the target device. Such differences are inevitable in practical SCA, but have often been disregarded in recent research. Therefore, the impact of device variation on cross-device SCA is analyzed. A method for Device Discrepancies Correction (DDC) is proposed, which can significantly reduce the discrepancy in sidechannel measurements caused by device variations. This is achieved by calculating the covariance shift between the source device and the target device, and then correcting the side-channel measurements of the target device accordingly. The approach is evaluated on eight Atmel XMEGA128A1U8 microcontrollers as a data pre-processing stage for cross-device SCA, and experimental results show that it can further improve the efficiency of existing cross-device SCA models.

关键词：

来源：评论

学校读者我要写书评

暂无评论

FedFC: An Efficient Personalized Federated Learning Method on Non-iid Data

FedFC: An Efficient Personalized Federated Learning Method o...

引用

International Conference on Parallel and Distributed Systems (ICPADS)

作者： Yuxin Yang Yifei Wang Chanying Huang Qianmu Li Shan Xiao School of Cyber Science and Engineering Nanjing University of Science and Technology Nanjing China School of Computer Science & Engineering Nanjing University of Science & Technology Nanjing China School of Cyber Science & Engineering Nanjing Univ. of Science & Technology Nanjing China Fiberhome R & D Center Fiberhome Communication Tech. Co. Wuhan China

Federated Learning has been widely used due to its ability to train models while ensuring data privacy and security. However, the presence of non-i.i.d. (independent and identically distributed) data among different participating entities leads to significant performance disparities in Federated Learning. In addition, as the essence of Federated Learning involves model training on local devices without centralized data storage, it poses certain challenges in personalized tasks. Traditional centralized machine learning methods can perform deep learning and personalized model training on centralized data, while distributed data in Federated Learning may not provide enough information to support personalized requirements. In this paper, we propose a personalized Federated Learning method called FedFC, which adopts parameter decoupling to address the data domain shift issue in Federated Learning. We experimentally validate our proposed solutions on the OrganCMNIST and COVID-19 datasets. The experimental results show that compared to existing methods, FedFC achieves average accuracy improvements of 4.93% and 6.31% on the two datasets, respectively. Additionally, by employing partial gradient uploading, we successfully reduce the communication overhead by 22.67% and 26.91% for each dataset.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Invited Paper: Actuator Trajectory Planning for UAVs with Overhead Manipulator using Reinforcement Learning

Invited Paper: Actuator Trajectory Planning for UAVs with Ov...

引用

IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC)

作者： Hazim Alzorgan Abolfazl Razi Ata Jahangir Moshayedi School of Computer Science Clemson University Clemson SC School of Information Engineering Jiangxi University of Sci. & Tech. Jiangxi China

In this paper, we investigate the operation of an aerial manipulator system, namely an Unmanned Aerial Vehicle (UAV) equipped with a controllable arm with two degrees of freedom to carry out actuation tasks on the fly. Our solution is based on employing a Q-learning method to control the trajectory of the tip of the arm, also called end-effector. More specifically, we develop a motion planning model based on Time To Collision (TTC), which enables a quadrotor UAV to navigate around obstacles while ensuring the manipulator’s reachability. Additionally, we utilize a model-based Q-learning model to independently track and control the desired trajectory of the manipulator’s end-effector, given an arbitrary baseline trajectory for the UAV platform. Such a combination enables a variety of actuation tasks such as high-altitude welding, structural monitoring and repair, battery replacement, gutter cleaning, sky scrapper cleaning, and power line maintenance in hard-to-reach and risky environments while retaining compatibility with flight control firmware. Our RL-based control mechanism results in a robust control strategy that can handle uncertainties in the motion of the UAV, offering promising performance. Specifically, our method achieves 92% accuracy in terms of average displacement error (i.e. the mean distance between the target and obtained trajectory points) using Q-learning with 15,000 episodes 1 .

关键词：

来源：评论

学校读者我要写书评

暂无评论

Learning Superconductivity from Ordered and Disordered Material Structures 38

Learning Superconductivity from Ordered and Disordered Mater...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Chen, Pin Peng, Luoxuan Jiao, Rui Mo, Qing Wang, Zhen Huang, Wenbing Liu, Yang Lu, Yutong National Supercomputer Center in Guangzhou School of Computer Science and Engineering Sun Yat-sen University China Dept. of Comp. Sci. & Tech. Institute for AI BNRist Center Tsinghua University China Institute for AIR Tsinghua University China Gaoling School of Artificial Intelligence Renmin University of China China Beijing Key Laboratory of Big Data Management and Analysis Methods Beijing China

Superconductivity is a fascinating phenomenon observed in certain materials under certain conditions. However, some critical aspects of it, such as the relationship between superconductivity and materials' chemical/structural features, still need to be understood. Recent successes of data-driven approaches in material science strongly inspire researchers to study this relationship with them, but a corresponding dataset is still lacking. Hence, we present a new dataset for data-driven approaches, namely SuperCon3D, containing both 3D crystal structures and experimental superconducting transition temperature (Tc) for the first time. Based on SuperCon3D, we propose two deep learning methods for designing high Tc superconductors. The first is SODNet, a novel equivariant graph attention model for screening known structures, which differs from existing models in incorporating both ordered and disordered geometric content. The second is a diffusion generative model DiffCSP-SC for creating new structures, which enables high Tc-targeted generation. Extensive experiments demonstrate that both our proposed dataset and models are advantageous for designing new high Tc superconducting candidates. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Efficient Registered Attribute Based Access Control with Same Sub-policies in Mobile Cloud Computing

引用

IEEE Transactions on Mobile Computing 2025年

作者： Weng, Wuwei Li, Jiguo Zhang, Yichen Lu, Yang Shen, Jian Han, Jinguang Fujian Normal University College of Computer and Cyber Security Fuzhou350117 China Fujian Fuzhou350117 China Fujian Provincial Key Laboratory of Network Security and Cryptology Fuzhou350117 China Nanjing Normal University School of Computer and Electronic Information Nanjing210023 China Zhejiang Sci-Tech University School of Information Science and Engineering Hangzhou310018 China Southeast University School of Cyber Science and Engineering Nanjing210096 China

Ciphertext-policy attribute-based encryption (CP-ABE) has long been considered as a promising access control tech.ology for cloud storage. However, CP-ABE depends on a central trusted authority to generate and distribute decryption keys, resulting in the key escrow issue. Most existing solutions only mitigate this problem but fail to resolve it entirely. Registered attribute-based encryption (RABE), a new cryptographic primitive, fundamentally addresses the key escrow problem by modifying the trust model, but its high computational overhead limits its practical application. Inspired by this challenge, we present an efficient registered attribute-based access control scheme designed for data encrypted with access policies containing the same sub-policy. In our scheme, users generate their own keys, while a key manager, who does not hold keys, replaces the central authority in managing users. Additionally, for data encrypted with the same sub-policy, the user's initial decryption stores the relevant parameters, which can be used for subsequent decryptions to reduce computational overhead. The proposed scheme is proven to achieve semantic security. Performance analysis demonstrates that our scheme enhances decryption efficiency by roughly 41.4% compared to existing RABE scheme, with a minimal storage trade-off, making it more practical for cloud storage application. © 2002-2012 IEEE.

关键词： Ciphertext

来源：评论

学校读者我要写书评

暂无评论

Actuator Trajectory Planning for UAVs with Overhead Manipulator using Reinforcement Learning

arXiv

引用

arXiv 2023年

作者： Alzorgan, Hazim Razi, Abolfazl Moshayedi, Ata Jahangir School of Computer Science Clemson University ClemsonSC United States School of Information Engineering Jiangxi University of Sci. & Tech. Jiangxi China

关键词： Antennas

来源：评论

学校读者我要写书评

暂无评论

BBDAS: Blockchain-Assisted Blinded Data Auditing Scheme for Cloud-Edge Systems 6th

BBDAS: Blockchain-Assisted Blinded Data Auditing Scheme for...

引用

6th International Conference on Blockchain and Trustworthy Systems, BlockSys 2024

作者： Wang, Chen Cui, Yuxin Yang, Bing Shen, Wei Tong, Wei School of Information Science and Engineering Zhejiang Sci-Tech University Hangzhou310018 China School of Computer Science Nanjing University of Information Science Technology Jiangsu210044 China Zhejiang Sci-Tech University Hangzhou310018 China Hangzhou Institute of Technology Xidian University Hangzhou311200 China

ISBN: (纸本)9789819614134

In distributed large-scale intelligent computing, efficient collaboration between the cloud and edge devices is essential to provide rapid data access for a vast number of users, which is crucial for optimal data storage services. The edge servers are utilized to meet the data access requirements of adjacent group users. Numerous data auditing schemes for cloud computing have been proposed to protect user data integrity and avoid the threat of data leakage, loss, or tampering. However, secure cloud-edge storage of group users remains to be further studied. This paper proposes a novel blockchain-assisted blinded data auditing scheme for cloud-edge systems. Firstly, blockchain tech.ology is utilized to monitor the activities of a semi-trusted third-party auditor (TPA). Secondly, user identifiers are utilized to trace the malicious behaviors in the systems. In addition, a data blinding algorithm is utilized to ensure that the raw data is unavailable to a curious TPA. Both security analysis and performance assessment demonstrate that this scheme conducts data auditing for cloud-edge systems with reliability and efficiency. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Data integrity

来源：评论

学校读者我要写书评

暂无评论

Investigating the Impact of Bug Dependencies on Bug-Fixing Time Prediction

Investigating the Impact of Bug Dependencies on Bug-Fixing T...

引用

International Symposium on Empirical Software engineering and Measurement, ESEM

作者： Chenglin Li Yangyang Zhao Yibiao Yang Yuming Zhou Liming Nie Zuohua Ding School of Computer Science and Technology Zhejiang Sci-Tech University Hangzhou China State Key Lab for Novel Software Technology Nanjing University Nanjing China School of Computer Science and Engineering Nanyang Technological University Singapore

Background: Bug dependencies refer to the link relationships between bugs and related issues, which are commonly observed in software evolution. It has been found that bugs with bug dependencies often take longer time to be resolved than other bugs without any dependencies. Despite the potential impact of bug dependencies on bug-fixing time, previous studies use traditional metrics without considering bug dependencies to build bug-fixing time prediction models. As a result, there is currently little empirical evidence to support the use of bug dependencies in improving prediction accuracy. Aims: We aim to conduct a comprehensive empirical study to investigate the value of considering bug dependencies for bug-fixing time prediction. Method: We define a set of bug dependency metrics based on bug dependencies. We first investigate the correlation between bug dependency metrics and bug-fixing time to investigate whether bugs with more complex dependencies are more time-consuming to be fixed. Next, we employ principal component analysis to study whether bug dependency metrics capture additional dimensions of a bug compared to traditional metrics. Finally, we build multivariate prediction models to explore whether considering bug dependencies can improve the effectiveness of bug-fixing time prediction. Results: The experimental results suggest that: (1) bugs with more complex dependencies require more time to be fixed; (2) bug dependency metrics are complementary to traditional metrics; (3) considering bug dependencies can improve the effectiveness of bug-fixing time prediction. Conclusions: These findings highlight the importance of considering bug dependencies in bug-fixing time prediction, and provide valuable insights into the potential impact of bug dependencies on software development processes.

关键词：

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：