检索结果-内蒙古大学图书馆

23rd USENIX Conference on File and Storage Technologies, FAST 2025

作者： Tian, Bing Liu, Haikun Tang, Yuhang Xiao, Shihai Duan, Zhuohui Liao, Xiaofei Jin, Hai Zhang, Xuecang Zhu, Junhua Zhang, Yu National Engineering Research Center for Big Data Technology and System Service Computing Technology and System Lab/Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology China Huawei Technologies Co. Ltd. China

ISBN: (纸本)9781939133458

Approximate nearest neighbor search (ANNS) has emerged as a crucial component of database and AI infrastructure. Ever-increasing vector datasets pose significant challenges in terms of performance, cost, and accuracy for ANNS services. None of modern ANNS systems can address these issues simultaneously. In this paper, we present FusionANNS, a high-throughput, low-latency, cost-efficient, and high-accuracy ANNS system for billion-scale datasets using SSDs and only one entry-level GPU. The key idea of FusionANNS lies in CPU/GPU collaborative filtering and re-ranking mechanisms, which significantly reduce I/O operations across CPUs, GPU, and SSDs to break through the I/O performance bottleneck. Specifically, we propose three novel designs: (1) multi-tiered indexing to avoid data swapping between CPUs and GPU, (2) heuristic re-ranking to eliminate unnecessary I/Os and computations while guaranteeing high accuracy, and (3) redundant-aware I/O deduplication to further improve I/O efficiency. We implement FusionANNS and compare it with the state-of-the-art SSD-based ANNS system-SPANN and GPU-accelerated in-memory ANNS system-RUMMY. Experimental results show that FusionANNS achieves 1) 9.4-13.1× higher query per second (QPS) and 5.7-8.8× higher cost efficiency compared with SPANN;2) and 2-4.9× higher QPS and 2.3-6.8× higher cost efficiency compared with RUMMY, while guaranteeing low latency and high accuracy. © 2025 FAST. All Rights Reserved.

关键词： Collaborative filtering

来源：评论

学校读者我要写书评

暂无评论

Mitigating Over-Unlearning in Machine Unlearning with Synthetic data Augmentation 24th

Mitigating Over-Unlearning in Machine Unlearning with Synt...

引用

24th International Conference on Algorithms and Architectures for Parallel Processing, ICA3PP 2024

作者： Wang, Baohai Qu, Youyang Gao, Longxiang Li, Conggai Li, Lin Smith, David Jinan China Shandong Provincial Key Laboratory of Computing Power Internet and Service Computing Shandong Fundamental Research Center for Computer Science Jinan China Data61 Commonwealth Science and Industry Research Organization CanberraVIC Australia College of Business and Law Royal Melbourne Institute of Technology MelbourneVIC Australia

ISBN: (纸本)9789819615445

In machine learning, data privacy and security has become an increasingly growing concern. The introduction of machine unlearning offers the ability to address this issue through the removal of personal and sensitive data from trained models to comply with laws, regulations, and user privacy requirements. However, despite the significant benefits of this technique, performing unlearning operations is only sometimes smooth in practice. When we attempt to remove specific data, the model may over-adjust, resulting in its performance on unlearned data. This phenomenon not only reduces the accuracy of the model on known data but also affects its ability to generalize new data, thus weakening the model’s overall performance. Therefore, in this paper, we analyze the phenomenon of over-unlearning;firstly, we explore how to mitigate the effects of over-unlearning by generating synthetic data to fill in the forgotten parts, using data synthesis-based techniques. Secondly, we combine the data synthesis-based compensation strategy with model fine-tuning to further improve the model’s adaptability and generalization capability by further training the model to accommodate synthetic data. Through comprehensive experimentation, we verify the effectiveness of the proposed data synthesis-based compensation strategy. The experimental results show that the data synthesis-based compensation strategy can effectively mitigate the effects of the over-unlearning phenomenon, as well as maintaining the stability and accuracy of the model after removing specific data. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： data assimilation

来源：评论

学校读者我要写书评

暂无评论

ELLIPTIC EQUATION WITH CRITICAL EXPONENT AND DIPOLE POTENTIAL: EXISTENCE AND DECAY ESTIMATES

引用

Acta Mathematica Scientia 2025年第2期45卷 636-658页

作者： Yu SU Zhisu LIU Senli LIU School of Mathematics and Big Data Anhui University of Science and TechnologyHuainan 232001China School of Mathematics and Physics China University of GeosciencesWuhan 430074China Institute for Advanced Marine Research China University of GeosciencesGuangzhou 511462China College of Mathematics and Computing Science Hunan University of Science and TechnologyXiangtan 411201China

The paper is concerned with a class of elliptic equation with critical exponent and Dipole *** precisely,we make use of the refined Sobolev inequality with Morrey norm to obtain the existence and decay properties of n... 详细信息

关键词： Dipole potential decay estimation Hardy Sobolev critical exponent Henon Sobolev critical exponent

来源：评论

学校读者我要写书评

暂无评论

Modal-Centric Insights Into Multimodal Federated Learning for Smart Healthcare: A Survey 24th

Modal-Centric Insights Into Multimodal Federated Learning fo...

引用

24th International Conference on Algorithms and Architectures for Parallel Processing, ICA3PP 2024

作者： Wang, Di Liu, Wenjian Gao, Longxiang Qu, Y. Neil. Zhang, Hu Shi, Jihong The Faculty of Data Science City University of Macau Macau China Jinan China Shandong Provincial Key Laboratory of Computing Power Internet and Service Computing Shandong Fundamental Research Center for Computer Science Jinan China Medical Integration and Practice Center Shandong University Jinan China

ISBN: (纸本)9789819615476

Federated Learning (FL) has progressed, providing a distributed mechanism where data need not be consolidated, thereby enhancing the privacy and security of sensitive healthcare data. Recent advancements in multimodal machine learning have empowered the development of accurate and robust AI systems in the medical domain, especially within centralized database systems. Multimodal federated learning (MFL) combines federated learning’s distributed nature with the flexibility to work with various data types. This capability is particularly relevant in medical contexts, where healthcare data can range from electronic health records (EHR) to radiology imaging, genomic data, and wearable device information. The key advantage of MFL in medical settings is the ability to create robust models without centralizing sensitive patient information. However, there are no existent research to comprehensively review the application status of MFL in smart healthcare systems from the modal-centric aspect. Motivated by this, we explore the systematization of knowledge about multimodal federated learning on medical data, find the challenges of this area and related solutions. With this research, we hope to provide a systematical review of this under-explored domain and provide insights for forthcoming researchers. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Towards practical data alignment in production federated learning

引用

Frontiers of Computer Science 2025年第1期19卷 167-169页

作者： Yexuan SHI Wei YU Yuanyuan ZHANG Chunbo XUE Yuxiang ZENG Zimu ZHOU Manxue GUO Lun XIN Wenjing NIE State Key Laboratory of Complex&Critical Software Environment and Advanced Innovation Center for Future Blockchain and Privacy Computing Beihang UniversityBeijing 100191China Zhongguancun Pan Connected Mobile Communication Technology Innovation and Application Research Institute Beijing 100088China China Mobile Research Institute Beijing 100053China School of Data Science City University of Hong KongHong Kong 999077China

1 Introduction Federated learning has emerged as a promising par-adigm for collaborative model training that facilitates cooperation among multiple parties while ensuring data privacy[1].Successful alignment of data across parties is crucial for effective federated learning[2].This alignment involves harmonizing heterogeneous data from different parties to identify shared data for joint model *** set intersection(PSI)is a technique that allows the alignment of common entities between parties without revealing additional ***,efficiently performing data alignment with PSI in federated learning[3],especially when dealing with highly unbalanced data,remains challenging due to the low efficiency.

关键词： alignment dealing parties

来源：评论

学校读者我要写书评

暂无评论

Machine learning approach to identity resolution for criminal profiling

引用

Journal of Cyber Security Technology 2025年

作者： Kazemian, Hassan Shrestha, Subeksha AI and Data Science Research Group School of Computing and Digital Media London Metropolitan University London United Kingdom

A common dilemma when working on criminal data is that often people manipulate their details to disguise themselves and hide their identities which leads to creating ambiguous and false identities. Deep Neural Network (DNN) is applied to work well on fraudulent and imbalanced data. Two subcategories of DNN, the Multilayer Perceptron (MLP) and Long Short-Term Memory (LSTM) with Python libraries such as TensorFlow and Keras are applied to the dataset for the detection of suspects with false identities to assist the process of analytical investigation by law enforcement agencies. Upon application of this approach to the anonymized policing dataset from SPIRIT Project funded by European Union’s Horizon, 5 main suspects with false identities were identified out of 23 targets on 39 million records. As working on criminal data is quite sensitive so to avoid data leakage while training and testing the data, K-fold validation has been applied. Furthermore, cascading both MLP and LSTM models on the policing dataset resulted in improved model prediction and accuracy compared to using each model individually. The cascaded model notably reduced false predictions for the recurring criminal patterns such as age-specific trends and most prevalent crime activities. © 2025 Informa UK Limited, trading as Taylor & Francis Group.

关键词： Information leakage

来源：评论

学校读者我要写书评

暂无评论

Few-Shot Source Separation for IoT Anti-Jamming via Multi-Task Learning and Meta-Learning

引用

IEEE Internet of Things Journal 2025年第11期12卷 16761-16776页

作者： Guo, Pengcheng Yu, Miao Li, Cheng An, Kang Sun, Yifu Chatzinotas, Symeon Niyato, Dusit National University of Defense Technology Sixty-Third Research Institute China State Key Laboratory of Complex Electromagnetic Environment Effects on Electronics and Information System China Kyung Hee University College of Electronics and Information Yongin-si17104 Korea Republic of Nanyang Technological University College of Computing and Data Science Singapore639798 Singapore

Malicious jamming attacks pose a significant threat to the integrity and performance of Internet of Things (IoT) networks. However, many jamming patterns are rare or infrequent, which makes them difficult to counter effectively. This paper addresses the critical issue of anti-jamming (AJ) under few-shot sample conditions in IoT networks. Source separation is a key component of AJ communication. Although deep learning-based source separation has demonstrated significant advantages, it typically requires a large amount of labeled data, which can be impractical in certain environments. To overcome this challenge, we propose two novel schemes that leverage multi-task learning (MTL) and meta-learning (ML) to enhance the model's signal separation capabilities within the constraints of limited sample scenarios. MTL enhances robustness by leveraging shared representations across tasks, while ML allows for rapid adaptation to novel jamming signals with minimal samples. Specifically, we employ a modified separation model, SepFormer, as our baseline and integrate MTL and ML schemes to enable the separation of unknown or few-shot jamming signals. Additionally, we have constructed two datasets encompassing both simulated and real-world environmental data to test and evaluate the performance of the proposed methods. Simulation results demonstrate the superior AJ performance of our schemes, particularly when compared with a direct application of the separation model with few-shot samples. Furthermore, our evaluation of performance across various jamming scenarios and interference-to-signal ratios (ISR) further confirms the effectiveness of our proposed scheme. © 2014 IEEE.

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

Seer: Accelerating Block chain Transaction Execution by Fine-Grained Branch Prediction 51st

Seer: Accelerating Block chain Transaction Execution by Fine...

引用

51st International Conference on Very Large data Bases, VLDB 2025

作者： Zhang, Shijie Cheng, Ru Liu, Xinpeng Xiao, Jiang Jin, Hai Li, Bo National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan430074 China Department of Computer Science and Engineering Hong Kong University of Science and Technology Hong Kong

Increasingly popular decentralized applications (dApps) with complex application logic incur significant overhead for executing smart contract transactions, which greatly limits public block chain performance. Pre-executing transactions off the critical path can mitigate substantial I/O and computation costs during execution. However, pre-execution does not yield any state transitions, rendering the system state inconsistent with actual execution. This inconsistency can lead to deviations in pre-execution paths when processing smart contracts with multiple state-related branches, thus diminishing pre-execution effectiveness. In this paper, we develop Seer, a novel public block chain execution engine that incorporates fine-grained branch prediction to fully exploit pre-execution effectiveness. Seerpredicts state-related branches using a two-level prediction approach, reducing inconsistent execution paths more efficiently than executing all possible branches. To enable effective reuse of pre execution results, Seer employs checkpoint-based fast-path execution, enhancing transaction execution for both successful and unsuccessful predictions. Evaluations with realistic block chain workloads demonstrate that Seer delivers an average of 27.7× transaction-level speedup and an overall 20.6× speedup in the execution phase over vanilla Ethereum, outperforming existing block chain execution acceleration solutions. © 2025, VLDB Endowment, All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

CAT-Net: A Co-Adaptive Transfer Learning Network for BCI-Assisted Neurorehabilitation

CAT-Net: A Co-Adaptive Transfer Learning Network for BCI-Ass...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Zhang, Shuailei Ding, Yi Jiang, Muyun Tang, Ning Chew, Effie Ang, Kai Keng Guan, Cuntai College of Computing and Data Science Nanyang Technological University Singapore National University Hospital Singapore Singapore National University Hospital National University of Singapore Singapore Agency for Science Technology and Research Nanyang Technological University Singapore

ISBN: (纸本)9798350368741

Brain-computer interfaces (BCIs) hold great potential for motor recovery in post-stroke patients. However, the motor imagery decoding accuracy is limited by the nonstationarity of EEG signals across subjects and sessions. We propose CAT-Net: a Co-Adaptive Transfer learning network to simultaneously address the inter-subject variability and intersession nonstationarity in EEG data. The proposed method selects a relevant subset of data from all the available subjects' data to train an initial model, followed by subject-specific transfer learning from the initial model to the target subject to establish a pretrain model. Subsequently, online adaptive training is then applied to incrementally train the pretrain model using the data from previous sessions for the target subject. This proposed network using this unique co-adaptive training method is then evaluated on both upper and lower-limb neurorehabilitation EEG datasets comprising 358 sessions from 33 stroke patients. The results showed significant accuracy improvements, achieving averaged accuracies of 70.6% and 72.3% on the respective datasets, surpassing the state-of-the-art baselines. © 2025 IEEE.

关键词： Electroencephalography Neurorehabilitation Transfer Learning

来源：评论

学校读者我要写书评

暂无评论

A Novel Hybrid Approach-Based on Heartbeat and Accelerometer Signals for Enhanced Security in WBSNs

引用

IEEE Sensors Journal 2025年第6期25卷 10295-10308页

作者： Yan, Guixiang Zhang, Guanghe Liu, Fanghua Williams Samuel, Oluwarotimi Aborokbah, Majed Guo, Jiquan Zhu, Liqing Wei, Sikang Jiangxi Normal University School of Computer and Information Engineering Nanchang330022 China University of Derby School of Computing and Data Science Research Centre DerbyDE22 3AW United Kingdom INTI International University Faculty of Data Science and Information Technology Nilai71800 Malaysia University of Tabuk Faculty of Computers and Information Technology Tabuk71491 Saudi Arabia Jiangxi Normal University School of Digital Industry Shangrao334000 China

In wireless body sensor networks (WBSNs), ensuring secure and efficient key distribution is critical, particularly given the limited computational and energy resources of the sensors. Existing methods often struggle to balance security with these resource constraints, especially in environments involving physiological data such as acceleration (ACC) and electrocardiogram (ECG) signals. For the first time, this study proposes a novel hybrid approach that integrates fuzzy commitment with ACC signal noise and ECG features for efficient key generation and distribution in WBSNs. By employing a low-pass filter to process ACC signal noise, we generated highly random binary sequences (BSs), leveraging the inherent randomness of the signal for secure key generation. Concurrently, an optimized coding scheme was built for ECG feature construction to ensure secure key distribution between devices. Extensive experiments, including entropy analysis and National Institute of Standards and Technology (NIST) statistical tests, confirm the robustness and security of our method. The proposed scheme achieves a false acceptance rate (FAR) of 5.23%, demonstrating superior performance across multiple databases in comparison to benchmark approaches. This novel dual-key generation strategy that combines the unpredictability of ACC noise with the individual-specific traits of ECG signals can significantly enhance the security, applicability, and versatility of WBSNs. © 2001-2012 IEEE.

关键词： Acceptance tests

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：