检索结果-内蒙古大学图书馆

CADEC: A Combinatorial Auction for Dynamic Distributed DNN Inference Scheduling in Edge-Cloud Networks

IEEE Transactions on Mobile computing 2025年

作者： Xu, Xiaolong Hu, Yuhao Cui, Guangming Qi, Lianyong Dou, Wanchun Nanjing University of Information Science and Technology School of Software Nanjing210044 China Nanjing University of Information Science and Technology Jiangsu Province Engineering Research Center of Advanced Computing and Intelligent Services School of Software China Nanjing University State Key Laboratory for Novel Software Technology China College of Computer science and technology Qingdao China Georgia State University Department of Computer Science United States

Deep Neural Network (DNN) Inference, as a key enabler of intelligent applications, is often computation-intensive and latency-sensitive. Combining the advantages of cloud computing (abundant computing resources) and edge computing (fast transmission), edge-cloud collaborative DNN inference is a powerful solution to these problems. However, in edge-cloud networks with heterogeneous resources, how to obtain reasonable decisions on server selection, model partition and resource allocation for efficient distributed DNN inference is a hard challenge. Furthermore, it is non-trivial to design suitable resource prices to maximize the social welfare. These challenges even escalate in dynamic edge-cloud networks where decisions should be generated as soon as each user arrives without future information. Therefore, we design a combinatorial auction for dynamic distributed DNN inference scheduling, named CADEC. CADEC first constructs a bid set for each user based on convex optimization theory for optimal solution searching. Next, prices of resources in the edge-cloud network are adjusted according to changes in supply-demand relationship, and whether to admit the request of each user is decided. Finally, the dynamic distributed inference scheduling decisions are generated through the primal-dual algorithm to maximize the social welfare. Theoretical analysis shows the good competitive ratio and polynomial time complexity of CADEC. Results of simulation experiments present that CADEC improves social welfare by up to 46% compared with state-of-the-art distributed DNN inference schemes. © 2002-2012 IEEE.

关键词： Convex optimization

来源：评论

学校读者我要写书评

暂无评论

Energy Efficient Collaborative Federated Learning Design: A Graph Neural Network based Approach

Energy Efficient Collaborative Federated Learning Design: A ...

引用

IEEE Conference on Global Communications (GLOBECOM)

作者： Nuocheng Yang Sihua Wang Mingzhe Chen Christopher G. Brinton Changchuan Yin Beijing Laboratory of Advanced Information Network Beijing University of Posts and Telecommunications Beijing China State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing China Department of Electrical and Computer Engineering Institute for Data Science and Computing University of Miami Coral Gables FL USA School of Electrical and Computer Engineering Purdue University West Lafayette IN USA

In this paper, we consider the design of an energy efficient collaborative federated learning (CFL) methodology where devices exchange their local FL parameters with a subset of their neighbors without reliance on a parameter server. In the considered model, mobile devices implement the designed CFL to train their local FL models using their own datasets over a realistic wireless network. Due to the limited wireless resources and user movements, each device may not be able to transmit its FL parameters with all neighboring devices. Therefore, each device must select a subset of devices to share its FL parameters and optimize the transmit power. This problem is formulated as an optimization problem, whose goal is to minimize CFL training energy consumption while satisfying the delay and CFL training loss requirements. To solve this problem, a two-stage solution is proposed. At the first stage, a graph neural network (GNN) based algorithm is proposed, which enables each device to individually determine the subset of devices to transmit FL parameters using its neighboring devices' location and connection information. Compared to standard iterative algorithms that need to iteratively optimize device connections and transmit power, the proposed GNN based method can directly obtain the optimal device connections without iterative optimization. Given the optimal device connections, at the second stage, each device can directly obtain the optimal transmit power. Simulation results show that the proposed algorithm can decrease energy consumption by up to 46% compared to the algorithm where each device will directly connect to its first and second nearest neighbors.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Alleviating Imbalance in Synchronous Distributed Training of Deep Neural Networks

Alleviating Imbalance in Synchronous Distributed Training of...

引用

IEEE International Conference on Big Data and Cloud computing (BdCloud)

作者： Haiyang Lin Mingyu Yan Duo Wang Wenming Li Xiaochun Ye Zhimin Tang Dongrui Fan State Key Laboratory of Computer Architecture ICT CAS Beijing China University of Chinese Academy of Sciences Beijing China State Key Laboratory of Mathematical Engineering and Advanced Computing Wuxi China

Deep neural networks achieve huge success in computer vision, natural language process and other fields. As larger networks and larger datasets bring better performance, the training time grows rapidly. Distributed training uses parallelization strategy to accelerate the training of deep neural network. In the existing distributed training framework of deep neural network, the parameter server architecture has poor scalability and is not suitable for large-scale distributed training. Ring/HD All Reduce algorithm perform well in scalability under ideal circumstances. However, in many realistic scenarios, imbalance will occur frequently in the distributed training system, which is mainly the imbalance of compute ability and network. Due to the structure characteristics of synchronous distributed training and All Reduce algorithm, imbalance will bring serious impact on performance. At present, there is a lack of research on both compute ability imbalance and network imbalance, especially dynamic imbalance. To alleviate this problem, we analyze All Reduce algorithm first, and propose RingDL, which optimizes the performance of distributed training system by alleviating imbalance through customizable dynamic workloads strategy and link strategy. The experimental results show that RingDL significantly reduces idle time in computing phase and synchronous phase, improving the overall performance.

关键词： Training Deep learning Degradation Computer vision Heuristic algorithms Scalability Neural networks

来源：评论

学校读者我要写书评

暂无评论

Streamline Ring ORAM Accesses through Spatial and Temporal Optimization

Streamline Ring ORAM Accesses through Spatial and Temporal O...

引用

IEEE Symposium on High-Performance Computer Architecture

作者： Dingyuan Cao Mingzhe Zhang Hang Lu Xiaochun Ye Dongrui Fan Yuezhi Che Rujia Wang Tsinghua University Beijing China State Key Laboratory of Computer Architecture ICT CAS Beijing China State Key Laboratory of Mathematical Engineering and Advanced Computing China Illinois Institute of Technology

Memory access patterns could leak temporal and spatial information in a sensitive program; therefore, obfuscated memory access patterns are desired from the security perspective. Oblivious RAM (ORAM) has been the favored candidate to eliminate the access pattern leakage through randomly remapping data blocks around the physical memory space. Meanwhile, accessing memory with ORAM protocols results in significant memory bandwidth overhead. For each memory request, after going through the ORAM obfuscation, the main memory needs to service tens of actual memory accesses, and only one real access out of them is useful for the program execution. Besides, to ensure the memory bus access patterns are indistinguishable, extra dummy blocks need to be stored and transmitted, which cause memory space waste and poor performance. In this work, we introduce a new framework, String ORAM, that accelerates the Ring ORAM accesses with Spatial and Temporal optimization schemes. First, we identify that dummy blocks could significantly waste memory space and propose a compact ORAM organization that leverages the real blocks in memory to obfuscate the memory access pattern. Then, we identify the inefficiency of current transaction-based Ring ORAM scheduling on DRAM devices and propose an effective scheduling technique that can overlap the time spent on row buffer misses while ensuring correctness and security. With a minimal modification on the hardware and software, and negligible impact on security, the framework reduces 30.05% execution time and up to 40% memory space overhead compared to the state-of-the-art bandwidth-efficient Ring ORAM.

关键词： Protocols Space debris Memory management Random access memory Performance gain Scheduling Software

来源：评论

学校读者我要写书评

暂无评论

Phase Reentrances and Solid Deformations in Confined Colloidal Crystals

引用

Physical Review Letters 2024年第1期132卷 018202-018202页

作者： Xiaoxia Li Huang Fang Krongtum Sankaewtong Minhuan Li Yanshuang Chen Jiping Huang Ran Ni Hajime Tanaka Peng Tan State Key Laboratory of Surface Physics and Department of Physics Fudan University Shanghai 200433 China Institute for Nanoelectronic Devices and Quantum Computing Fudan University Shanghai 200433 China Chemical Engineering School of Chemical and Biomedical Engineering Nanyang Technological University 62 Nanyang Drive 637459 Singapore Department of Fundamental Engineering Institute of Industrial Science University of Tokyo 4-6-1 Komaba Meguro-ku Tokyo 153-8505 Japan Research Center for Advanced Science and Technology University of Tokyo 4-6-1 Komaba Meguro-ku Tokyo 153-8904 Japan

A simple geometric constraint often leads to novel, complex crystalline phases distinct from the bulk. Using thin-film charge colloidal crystals, a model system with tunable interactions, we study the effects of geometric constraints. Through a combination of experiments and simulations, we systematically explore phase reentrances and solid deformation modes concerning geometrical confinement strength, identifying two distinct categories of phase reentrances below a characteristic layer number, Nc: one for bcc bulk-stable and another for fcc bulk-stable systems. We further verify that the dominant thermodynamic origin is the nonmonotonic dependence of solids’ free energy on the degree of spatial confinement. Moreover, we discover transitions in solid deformation modes between interface-energy and bulk-energy dominance: below a specific layer number, Nk, geometric constraints generate unique soft deformation modes adaptive to confinement. These findings on the N-dependent thermodynamic and kinetic behaviors offer fresh insights into understanding and manipulating thin-film crystal structures.

关键词： Phase transitions by order Colloidal crystal Colloids

来源：评论

学校读者我要写书评

暂无评论

A Security Sensitive Function Mining Approach Based on Precondition Pattern Analysis

引用

Computers, Materials & Continua 2020年第5期63卷 1013-1029页

作者： Zhongxu Yin Yiran Song Huiqin Chen Yan Cao State Key Laboratory of Mathematical Engineering and Advanced Computing Zhengzhou450001China Henan University of Animal Husbandry&Economy Zhengzhou450046China University of Michigan Transportation Research Institute Michigan48109-2150USA Zhengzhou University Zhengzhou450001China

Security-sensitive functions are the basis for building a taint-style vulnerability *** approaches for extracting security-sensitive functions either don’t analyze data flow accurately,or not conducting pattern analyzing of conditions,resulting in higher false positive rate or false negative rate,which increased manual confirmation *** this paper,we propose a security sensitive function mining approach based on preconditon pattern ***,we propose an enhanced system dependency graph analysis algorithm for precisely extracting the conditional statements which check the function parameters and conducting statistical analysis of the conditional statements for selecting candidate security sensitive functions of the target *** we adopt a precondition pattern mining method based on conditional statements nomalizing and *** with fixed precondition patterns are regarded as security-sensitive *** experimental results on four popular open source codebases of different scales show that the approach proposed is effective in reducing the false positive rate and false negative rate for detecting security sensitive functions.

关键词： Code mining security sensitive function function preconditions single-linkage clustering

来源：评论

学校读者我要写书评

暂无评论

Medical Image Registration and Its Application in Retinal Images: A Review

arXiv

引用

arXiv 2024年

作者： Nie, Qiushi Zhang, Xiaoqing Hu, Yan Gong, Mingdao Liu, Jiang Research Institute of Trustworthy Autonomous Systems Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen518055 China Center for High Performance Computing Shenzhen Key Laboratory of Intelligent Bioinformatics Shenzhen institute of Advanced Technology Chinese Academy of Sciences Shenzhen518055 China Singapore Eye Research Institute 169856 Singapore State Key Laboratory of Ophthalmology Optometry and Visual Science Eye Hospital Wenzhou Medical University Wenzhou325027 China

Medical image registration is vital for disease diagnosis and treatment with its ability to merge diverse information of images, which may be captured under different times, angles, or modalities. Although several surveys have reviewed the development of medical image registration, these surveys have not systematically summarized methodologies of existing medical image registration methods. To this end, we provide a comprehensive review of these methods from traditional and deep learning-based directions, aiming to help audiences understand the development of medical image registration quickly. In particular, we review recent advances in retinal image registration at the end of each section, which has not attracted much attention. Additionally, we also discuss the current challenges of retinal image registration and provide insights and prospects for future research. © 2024, CC BY-NC-SA.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Ultrabright Multiplexed Energy-Time-Entangled Photon Generation from Lithium Niobate on Insulator Chip

引用

Physical Review Applied 2021年第6期15卷 064059-064059页

作者： Guang-Tai Xue Yun-Fei Niu Xiaoyue Liu Jia-Chen Duan Wenjun Chen Ying Pan Kunpeng Jia Xiaohan Wang Hua-Ying Liu Yong Zhang Ping Xu Gang Zhao Xinlun Cai Yan-Xiao Gong Xiaopeng Hu Zhenda Xie Shining Zhu National Laboratory of Solid State Microstructures School of Physics School of Electronic Science and Engineering College of Engineering and Applied Sciences and Collaborative Innovation Center of Advanced Microstructures Nanjing University Nanjing 210093 China State Key Laboratory of Optoelectronic Materials and Technologies and School of Physics and Engineering Sun Yat-sen University Guangzhou 510275 China Institute for Quantum Information and State Key Laboratory of High Performance Computing College of Computing National University of Defense Technology Changsha 410073 China

A high-flux entangled-photon source is a key resource for quantum optical study and application. Here, it is realized in a lithium niobate on isolator (LNOI) chip, with 2.79 × 1011 Hz/mW photon-pair rate and 1.53 × 109 Hz/(nm mW) spectral brightness. These data are boosted by over 2 orders of magnitude compared with existing technologies. A 160-nm-broad bandwidth is engineered for eight-channel multiplexed energy-time entanglement. Harnessed by high-extinction-frequency correlation and Franson interferences up to 99.17% visibility, such energy-time-entanglement multiplexing further enhances the high-flux data rate and warrants broad application in quantum-information processing on a chip.

关键词： Integrated optics Optical quantum information processing Optoelectronics Photon pairs & parametric down-conversion Quantum state engineering Second order nonlinear optical processes Optical sources & detectors Waveguides

来源：评论

学校读者我要写书评

暂无评论

AttCorr: A Novel Deep Learning Model for Flow Correlation Attacks on Tor

AttCorr: A Novel Deep Learning Model for Flow Correlation At...

引用

Consumer Electronics and Computer engineering (ICCECE), 2021 IEEE International Conference on

作者： Ji Li Chunxiang Gu Xieli Zhang Xi Chen Wenfen Liu State Key Laboratory of Mathematical Engineering and Advanced Computing Zhengzhou China School of Guangxi Key Laboratory of Cryptography and Information Security Guilin University of Electronic Technology Guilin China

ISBN: (数字)9781728183190

ISBN: (纸本)9781728183206

The proliferation of illegal information and criminal behavior on anonymous networks arouses the demand for deanonymization attacks on anonymous communication systems. Flow correlation is a common technique for deanonymization attacks, but most related works currently are based on statistical learning models, thus suffering from burdensome artificial feature engineering. In this paper, we design a novel deep learning model AttCorr for flow correlation attacks on Tor. AttCorr takes the raw traffic features of flow pair as input, including packet sizes, flow directions and inter-packet delays, and uses the multi-head attention mechanism to capture the flow information involved in features and the complex nature of noise in Tor. Experiment results show that AttCorr achieves the same level of accuracy as the stateof-the art method of DeepCorr with lower complexity, simpler feature processing and better interpretability.

关键词： Deep learning Training Correlation Statistical learning Telecommunication traffic Feature extraction Delays

来源：评论

学校读者我要写书评

暂无评论

A Novel Deep Learning Approach Featuring Graph-Based Algorithm for Cell Segmentation and Tracking

A Novel Deep Learning Approach Featuring Graph-Based Algorit...

引用

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Keliang Zhao Jovial Niyogisubizo Linxia Xiao Yi Pan Dongqing Wei Didi Rosiyadi Yanjie Wei Shenzhen Key Laboratory of Intelligent Bioinformatics and Center for High Performance Computing Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China College of Computer Science and Control Engineering Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Shenzhen China State Key Laboratory of Microbial Metabolism and School of Life Sciences and Biotechnology Shanghai Jiao Tong University Shanghai China Research Center for Artificial Intelligence and Cybersecurity National Research and Innovation Agency Bandung Indonesia

The precise segmentation and tracking of cells in microscopy image sequences play a pivotal role in biomedical research, facilitating the study of tissue, organ, and organism development. However, manual segmentation and tracking of cells is time-consuming and often require professional experiences. Besides, segmenting cells in the images with a low signal-to-noise ratio remains difficult. While deep learning (DL) has become a common method for cell segmentation, few DL- based methods address concurrent cell segmentation and tracking. In this paper, we propose a novel DL approach featuring graph-based tracking for cell segmentation and tracking in microscopy images. We combine Deeplabv3+ for semantic segmentation and ResNet50 for enhanced feature extraction, enabling comprehensive cell detection and instance segmentation. Post-processing, involving non-maxima suppression and outlier detection, refines predictions and produces final segmentation. The tracking method is based on the relative position of graph nodes to track segmented cells, encompassing cell division and apoptosis. We conduct our experiments on the induced pluripotent stem (iPS) cell datasets, and the results show that the segmentation and tracking performance of our method yields superior performance compared to the benchmark models. More specifically, our approach achieved DET values of 0.955 and 0.913, TRA values of 0.951 and 0.906, and SEG values of 0.690 and 0.665 on two iPS dataset videos, respectively. Additionally, the performance assessment encompasses four real microscopy datasets from the Cell Tracking Challenge (CTC). Our method greatly reduces the cost of manual labeling and labor-intensive costs.

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：