检索结果-内蒙古大学图书馆

CMD: A Cache-Assisted GPU Memory Deduplication Architecture

IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems 2025年

作者： Zhao, Wei Feng, Dan Tong, Wei Wei, Xueliang Wu, Bing Ministry of Education of China School of Computer Science and Technology Huazhong University of Science and Technology Wuhan National Laboratory for Optoelectronics Key Laboratory of Information Storage System Engineering Research Center of data storage systems and Technology China

Massive off-chip accesses in GPUs are the main performance bottleneck. We find that many writes are duplicate, and the duplication can be inter-dup and intra-dup. While inter-dup means different memory blocks are identical, and intra-dup means all the 4B elements in a line are the same. In this work, we propose a cache-assisted GPU memory deduplication architecture named CMD to reduce the off-chip accesses via utilizing the data duplication in GPU applications. CMD includes three key design contributions which aim to reduce the three kinds of accesses: (1) A novel GPU memory deduplication architecture that removes the intra-dup and inter-dup lines. We design several techniques to manage duplicate blocks, reducing massive off-chip writes. (2) We propose a cache-assisted read scheme to reduce the reads to duplicate data. When an L2 cache miss wants to read the duplicate block, if the reference block has been fetched to L2 and it is clean, we can copy it to the L2 missed block without accessing off-chip DRAM. As for the reads to intra-dup data, CMD uses the on-chip metadata cache to get the data. (3) When a cache line is evicted, the clean sectors in the line are invalidated while the dirty sectors are written back. However, most read-only victims are re-referenced from DRAM more than twice. Therefore, we add a full-associate FIFO to accommodate the read-only (it is also clean) victims to reduce the re-reference counts. Experiments show that CMD can decrease the off-chip accesses by 31.01%, reduce the energy by 32.78% and improve performance by 42.53%. Besides, CMD can improve the performance of memory-intensive workloads by 57.56%. © 2025 IEEE.

关键词： Cache memory

来源：评论

学校读者我要写书评

暂无评论

KAN v.s. MLP for Offline Reinforcement Learning

KAN v.s. MLP for Offline Reinforcement Learning

引用

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

作者： Haihong Guo Fengxin Li Jiao Li Hongyan Liu School of Information Renmin University of China China Institute of Medical Information / Medical Library Chinese Academy of Medical Sciences / Peking Union Medical College China Key Laboratory of Data Engineering and Knowledge Engineering Ministry of Education China School of Economics and Management Tsinghua University China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Kolmogorov-Arnold Networks (KAN) is an emerging neural network architecture in machine learning. It has greatly interested the research community about whether KAN can be a promising alternative to the commonly used Multi-Layer Perceptions (MLP). Experiments in various fields demonstrated that KAN-based machine learning can achieve comparable if not better performance than MLP-based methods, but with much smaller parameter scales and are more explainable. In this paper, we explore the incorporation of KAN into the actor and critic networks for offline reinforcement learning (RL). We evaluated the performance, parameter scales, and training efficiency of various KAN and MLP-based conservative Q-learning (CQL) on the classical D4RL benchmark for offline RL. Our study demonstrates that KAN can achieve performance close to the commonly used MLP with significantly fewer parameters. This allows us to choose the base networks according to the offline RL task requirements.

关键词： Training Performance evaluation Hands Q-learning Neural networks Memory management Signal processing Benchmark testing Splines (mathematics) Speech processing

来源：评论

学校读者我要写书评

暂无评论

Convolution Type of Metaplectic Cohen's Distribution Time-Frequency Analysis Theory, Method and Technology

引用

IEEE Transactions on Information Theory 2025年第3期71卷 2292-2314页

作者： Cui, Manjun Zhang, Zhichao Han, Jie Chen, Yunjie Cao, Chunzheng Nanjing University of Information Science and Technology School of Mathematics and Statistics Center for Applied Mathematics of Jiangsu Province Jiangsu International Joint Laboratory on System Modeling and Data Analysis Nanjing210044 China Nanjing University of Information Science and Technology School of Mathematics and Statistics Nanjing210044 China Shanghai Jiao Tong University Key Laboratory of System Control and Information Processing Ministry of Education Shanghai200240 China Hainan Normal University Key Laboratory of Computational Science and Application of Hainan Province Haikou571158 China Nanjing University of Information Science and Technology School of Remote Sensing and Geomatics Engineering Nanjing210044 China

The conventional Cohen's distribution can't meet the requirement of additive noises jamming signals high-performance denoising under the condition of low signal-to-noise ratio, it is necessary to integrate the metaplectic transform for non-stationary signal fractional domain time-frequency analysis. In this paper, we embark on blending time-frequency operators and coordinate operator fractionizations to formulate the definition of the joint fractionizations metaplectic Wigner distribution (JFMWD), based on which we integrate the generalized metaplectic convolution to address the unified representation issue of the convolution type of metaplectic Cohen's distribution (CMCD), whose special cases and essential properties are also derived. We blend Wiener filter principle and fractional domain filter mechanism of the metaplectic transform to design the least-squares adaptive filter method in the JFMWD domain, giving birth to the least-squares adaptive filter-based CMCD whose kernel function can be adjusted with the input signal automatically to achieve the minimum mean-square error (MSE) denoising in Wigner distribution domain. We discuss the optimal symplectic matrices selection strategy of the proposed adaptive CMCD through the minimum MSE minimization modeling and solving. Some examples are also carried out to demonstrate that the proposed filtering method outperforms some state-of-the-arts including the classical Gaussian filter, the Cohen's distribution filtering methods with fixed kernel functions, the classical Wiener filter, the adaptive generalized linear canonical convolution-based filtering method, the adaptive Cohen's distribution filtering method, the adaptive JFMWD-based Cohen's distribution filtering method, and the adaptive generalized metaplectic convolution-based Cohen's distribution filtering method in noise suppression. © 1963-2012 IEEE.

关键词： Adaptive filtering

来源：评论

学校读者我要写书评

暂无评论

VECLLF: A vehicle-edge collaborative lifelong learning framework for anomaly detection in VANETs

引用

Computer Networks 2025年 265卷

作者： Wang, Yingqing Liang, Yanhua Huang, Yue Qin, Guihe Jilin University College of Computer Science and Technology Changchun130012 China Jilin University Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Changchun130012 China Jilin University Center for Computer Fundamental Education Changchun130012 China

With the rapid development of intelligent connected vehicles, Vehicular Ad Hoc Network (VANET) is widely used in intelligent transportation systems. VANETs are vulnerable to attacks in their operating environments, such as denial of service (DoS) attacks and injection attacks. Thus, the development of an efficient anomaly detection framework is a promising solution. Most existing methods can only detect known attacks, are less efficient in detecting unknown attacks, and fail to deal with them. To solve these problems, the lifelong learning framework (LLF) is introduced, which aims to store and label suspicious data by applying blockchain and then retrain the model. However, existing LLFs are overly idealized as they do not adequately consider the key issues of label budget and storage space. To improve the practicality of existing methods, this paper proposes a novel LLF based on the existing LLF, called Vehicle Edge Collaborative Lifelong Learning Framework (VECLLF), which can deploy the model update task on the edge server. Specifically, the proposed anomaly detection module based on ensemble active learning is used to achieve high-performance detection under a limited label budget, the proposed incremental learning module based on a sample library continuously learns the model by replaying representative samples within limited storage space, the proposed optimization-based feature selection module is used to remove redundant and interfering features, and the explainable artificial intelligence (XAI) module is used to explain the model to ensure its interpretability. In addition, the blockchain-based database (BC-DB) module is used to help store suspicious data and interact with security agencies to label it. We conducted experiments on two public datasets, AWID2 and CICIDS2017, and the results show that VECLLF outperforms state-of-the-art anomaly detection models and existing LLF, with F1-scores exceeding 97% on both datasets. Furthermore, we collect data from th

关键词： Rapid transit

来源：评论

学校读者我要写书评

暂无评论

Retinex Variational Underwater Image Enhancement with Hyper-Laplacian and Edge-Preserving Filtering Priors 24

Retinex Variational Underwater Image Enhancement with Hyper-...

引用

Proceedings of the 2024 2nd Asia Symposium on Image and Graphics

作者： Xinheng Zhang Peixian Zhuang Junjie Tong Chenyu Wang Haojie Shi Weizhen Wang School of Automation and Electrical Engineering University of Science and Technology Beijing Beijing China The Key Laboratory of Knowledge Automation for Industrial Processes Ministry of Education School of Automation and Electrical Engineering University of Science and Technology Beijing Beijing China

ISBN: (纸本)9798400709906

To address the issues of low contrast, loss of structures and details, artifacts and noise in degraded underwater images, we propose a novel Retinex variational underwater image enhancement method with hyper-Laplacian and edge-preserving filter priors. Initially, a statistical-based technique is first utilized to correct color casts, followed by converting an underwater image from RGB to HSV color space. According to the Retinex theory, the value channel is decomposed into the illumination and reflectance. In our model, the L1/2-norm is imposed on the first-order gradients of both the illumination and reflectance to preserve structures and details of the underwater image. Moreover, an edge-preserving filtering prior is employed to constrain both the illumination and reflectance, eliminating undesirable artifacts and noise. Besides, the L2-norm is simple and effective to constrain the fidelity term. Next, an alternating direction strategy is implemented to optimize this objective. Finally, the enhanced illumination undergoes gamma correction, and the enhanced reflectance is merged to produce the improved value channel. The final enhanced underwater image is obtained by converting the image from HSV space to RGB. Numerous experiments demonstrates that our method excels in enhancing structures and details, as well as removing artifacts and noise on various underwater images. Furthermore, this approach can be extended to applications, such as dehazing and sandstorm removal, showcasing its versatility and effectiveness.

关键词： Alternating optimization

来源：评论

学校读者我要写书评

暂无评论

Predictive Beamforming Using Feature Fusion on Incomplete Time-Series Multimodal Sensing data

Predictive Beamforming Using Feature Fusion on Incomplete Ti...

引用

IEEE Conference on Wireless Communications and Networking

作者： Minghao Gao Xing Zhang Ronghui Zhang Xiaojun Jing School of Information and Communication Engineering Beijing University of Posts and Telecommunications Beijing China Key Laboratory of Knowledge Automation for Industrial Processes of Ministry of Education School of Automation and Electrical Engineering University of Science and Technology Beijing Beijing China

ISBN: (数字)9798350368369

ISBN: (纸本)9798350368376

Next-generation communications demand high-speed data transmission, making millimeter-wave (mmWave) and terahertz (THz) technologies essential. Reducing beam-forming overhead has led to the integration of sensing-assisted predictive beamforming with deep learning. However, in high-mobility scenarios, real-time collection of complete time-series data is often challenging due to network instability and device limitations. Motivated by this, we propose an LSTM-based method for completing incomplete time-series sensing data and develop a multimodal sensing-assisted predictive beamforming model. Experimental results show that the position data imputation error remains within meters, and training with the completed data improves model performance, achieving over 50% Top-1 and 93.53% Top-5 accuracy. This approach enhances prediction accuracy while reducing beamforming overhead

关键词： Meters Deep learning Accuracy Array signal processing Predictive models Feature extraction data models Imputation Long short term memory Global Positioning System

来源：评论

学校读者我要写书评

暂无评论

A Fast Framework for Efficiently Constructing Valuable Cubes 26th

A Fast Framework for Efficiently Constructing Valuable Cubes

引用

26th International Conference on Information and Communications Security, ICICS 2024

作者： Li, Jingtao Gao, Bo Wan, Jianxiong Li, Leixiao Zhang, Jiaxiang School of Data Science and Application Inner Mongolia University of Technology Hohhot China Inner Mongolia Key Laboratory of Beijiang Cyberspace Security Hohhot China School of Information Engineering Beijing Institute of Graphic Communication Beijing China Research Center of Large-Scale Energy Storage Technologies Ministry of Education of the People’s Republic of China Beijing China

ISBN: (纸本)9789819787975

The cube attack is a powerful technique for attacking symmetric ciphers, of which one of the core parts is the search for good cubes. The construction of candidate cubes targeting linear superpolies, proposed at Asiacrypt 2021 [20], is the dominant strategy for searching good cubes. However, the time spent on constructing candidate cubes increases significantly with initialization rounds on Trivium. To address this issue, we propose a novel framework called fast framework (The source code and data for this paper are available at https://***/Elpsys/Fast-Framework.), which enhances the efficiency of constructing candidates and discovers better cubes. The integration of numeric mapping and division property is the core of our framework;they can improve efficiency while guaranteeing accuracy. The proposed method is successfully applied to 830-and 832-round Trivium, both identifying more than 20 valuable cubes whose superpolies contain linear-independent terms and have the potential to be used in further practical key-recovery attacks. The experimental results demonstrated that the efficiency of constructing the mother cube is improved by at least a factor of 9 compared to conventional methods. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Geometry

来源：评论

学校读者我要写书评

暂无评论

Dual-Network Cross-Learning for Metabolite-Disease Association Prediction

IEEE Transactions on Computational Biology and Bioinformatic...

引用

IEEE Transactions on Computational Biology and Bioinformatics 2025年第2期22卷 545-556页

作者： Yanxin Chen Qiao Ning Yitong Zhang Hui Li Shikai Guo Department of Information Science and Technology Dalian Maritime University Dalian China School of Artificial Intelligence and Computer Science Jiangnan University Wuxi China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun China Dalian Key Laboratory of Artificial Intelligence Dalian China

In recent years, increasing evidence has demonstrated a close association between metabolites and various complex human diseases, providing valuable insights for disease diagnosis, treatment, and prevention. Although deep learning-based approaches have achieved certain success in predicting metabolic disease associations, challenges remain in enriching graph information and effectively integrating metabolic and disease features. To address these issues, this paper proposes a model named DCMDA, which extracts deep features of both metabolites and diseases using Dual-network Cross-learning for Metabolite- Disease Association prediction. DCMDA consists of three parts. The data processing module integrates similarity networks with association networks to construct a heterogeneous network. The feature extraction module extracts features from the metabolite-disease association network based on the non-negative matrix factorization method and from the heterogeneous network using graph autoencoder techniques. The feature fusion module combines the association matrix feature with the heterogeneous network feature through a Cross-Attention mechanism, thereby obtaining deep representations of metabolites and diseases. These features are then used to train the model to predict association scores between metabolites and diseases. Experimental results demonstrate that in 5-fold cross-validation, DCMDA achieves an area under the receiver operating characteristic curve (AUC) of 97.8% and an area under the precision-recall curve (AUPR) of 97.9%, outperforming state-of-the-art prediction methods.

关键词： Diseases Feature extraction data mining Semantics Deep learning Kernel Information entropy Prediction algorithms Noise Nearest neighbor methods

来源：评论

学校读者我要写书评

暂无评论

Robustifying vision transformer for image forgery localization with multi-exit architectures

引用

Pattern Recognition 2025年 164卷

作者： Shi, Zenan Chen, Haipeng Zhang, Dong College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun130012 China Department of Electronic and Computer Engineering The Hong Kong University of Science and Technology 999077 Hong Kong

The proliferation of image manipulation tools has led to an increase in the number of manipulated images being disseminated online, posing risks like the propagation of fake news and telecom fraud. Thus, there is an increasing demand for precise, generic, and robust methods for detecting and locating manipulated images. In this paper, we propose a simple and clean model, named MEAFormer, for image forgery localization that does not heavily rely on pre-trained models. MEAFormer comprises three main components: an encoder network, a neck network, and a decoder network. Specifically, the transformer-based encoder network extracts hierarchical feature representations from the input image, providing rich contextual information in each layer. The neck network, incorporating our proposed cross-layer feature aggregation (CFA), aggregates these hierarchical features. To achieve better spatial feature co-occurrence, instead of using noise or edge artifacts, we introduce a multi-scale graph reasoning (MGR) module within the decoder network via bipartite graphs over the encoder and decoder features in a multi-scale fashion. The cross-level enhancement (CLE) further performs adjacent-level feature fusion to amplify the regions of interest in aggregated manipulation features. Finally, the multi-exit architecture (MEA) guides the model to learn fine-grained features and segment out the manipulated region. Extensive experiments across diverse and challenging datasets conclusively establish the superiority of MEAFormer over existing state-of-the-art methods, excelling in accuracy, generalization, and robustness. © 2025 Elsevier Ltd

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

AHMSA-Net: Adaptive Hierarchical Multi-Scale Attention Network for Micro-Expression Recognition

arXiv

引用

arXiv 2025年

作者： Zhang, Lijun Zhang, Yifan Tang, Weicheng Sun, Xinzhi Wang, Xiaomeng Li, Zhanshan College of Computer Science and Technology Jilin University Changchun Jilin130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun Jilin130012 China

Micro-expression recognition (MER) presents a significant challenge due to the transient and subtle nature of the motion changes involved. In recent years, deep learning methods based on attention mechanisms have made some breakthroughs in MER. However, these methods still suffer from the limitations of insufficient feature capture and poor dynamic adaptation when coping with the instantaneous subtle movement changes of micro-expressions. Therefore, in this paper, we design an Adaptive Hierarchical Multi-Scale Attention Network (AHMSA-Net) for MER. Specifically, we first utilize the onset and apex frames of the micro-expression sequence to extract threedimensional (3D) optical flow maps, including horizontal optical flow, vertical optical flow, and optical flow strain. Subsequently, the optical flow feature maps are inputted into AHMSA-Net, which consists of two parts: an adaptive hierarchical framework and a multi-scale attention mechanism. Based on the adaptive downsampling hierarchical attention framework, AHMSA-Net captures the subtle changes of micro-expressions from different granularities (fine and coarse) by dynamically adjusting the size of the optical flow feature map at each layer. Based on the multi-scale attention mechanism, AHMSA-Net learns micro-expression action information by fusing features from different scales (channel and spatial). These two modules work together to comprehensively improve the accuracy of MER. Additionally, rigorous experiments demonstrate that the proposed method achieves competitive results on major micro-expression databases, with AHMSA-Net achieving recognition accuracy of up to 78.21% on composite databases (SMIC, SAMM, CASMEII) and 77.08% on the CASME3 database. Copyright © 2025, The Authors. All rights reserved.

关键词： Deep learning

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：