检索结果-内蒙古大学图书馆

2nd International Conference on Electronic Information Engineering and computer Communication, EIECC 2022

作者： Gao, Han Xu, Ling Zhao, Yixuan Yan, Tanbao Aviation Key Laboratory of Science and Technology on Airborene and Missile-borne Computer.Xi'an Aeronautics Computing Technology Research Institute Xi'an China

ISBN: (纸本)9781510663121

Most tasks in airborne scenarios are composed of neural networks (NN) and non-NN tasks. On the one hand, Deep neural networks (DNNs) have become the mainstream method of AI task processing. DNNs is accompanied by a large number of complex tensor calculations, as well as network delays and data privacy risks caused by using the cloud to process data. The combination of dedicated AI hardware accelerators and edge computing has become a new solution. On the other hand, dedicated AI hardware accelerators cannot efficiently handle non-NN class tasks (such as signal processing, control flow, etc.), so onboard systems need to design new hardware, complete compiler stacks, and runtime systems to handle non-NN class tasks. In order to effectively deal with multiple tasks in the airborne environment, the airborne computing system needs to establish a set of unified compilation and runtime system, which can optimize multiple tasks and be deployed on the heterogeneous computing platform to achieve "one compilation, multiple deployment". This paper analyzes the development of compiler and runtime systems at home and abroad, and compares the advantages and disadvantages of different technologies. Finally, this paper discusses the application scenarios of the compilation system and runtime in the airborne environment and the technical directions that need to be breakthroughs in the future, so as to provide guidance for the development of the airborne intelligent compilation and runtime system. © 2023 SPIE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

In-Context Compositional Generalization for Large Vision-Language Models

In-Context Compositional Generalization for Large Vision-Lan...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Li, Chuanhao Jing, Chenchen Li, Zhen Zhai, Mingliang Wu, Yuwei Jia, Yunde Beijing Key Laboratory of Intelligent Information Technology School of Computer Science & Technology Beijing Institute of Technology China Guangdong Laboratory of Machine Perception and Intelligent Computing Shenzhen MSU-BIT University China School of Computer Science Zhejiang University Hangzhou China

ISBN: (纸本)9798891761643

Recent work has revealed that in-context learning for large language models exhibits compositional generalization capacity, which can be enhanced by selecting in-context demonstrations similar to test cases to provide contextual information. However, how to exhibit in-context compositional generalization (ICCG) of large vision-language models (LVLMs) is non-trival. Due to the inherent asymmetry between visual and linguistic modalities, ICCG in LVLMs faces an inevitable challenge-redundant information on the visual modality. The redundant information affects in-context learning from two aspects: (1) Similarity calculation may be dominated by redundant information, resulting in sub-optimal demonstration selection. (2) Redundant information in in-context demonstrations brings misleading contextual information to in-context learning. To alleviate these problems, we propose a demonstration selection method to achieve ICCG for LVLMs, by considering two key factors of demonstrations: content and structure, from a multimodal perspective. Specifically, we design a diversity-coverage-based matching score to select demonstrations with maximum coverage, and avoid selecting demonstrations with redundant information via their content redundancy and structural complexity. We build a GQA-ICCG dataset to simulate the ICCG setting, and conduct experiments on GQA-ICCG and the VQA v2 dataset. Experimental results demonstrate the effectiveness of our method. © 2024 Association for Computational Linguistics.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

TransGPerf:Exploiting Transfer Learning for Modeling Distributed Graph Computation Performance

引用

Journal of computer Science & technology 2021年第4期36卷 778-791页

作者： Songjie Niu Shimin Chen State Key Laboratory of Computer Architecture Institute of Computing TechnologyChinese Academy of Sciences Beijing 100190China University of Chinese Academy of Sciences Beijing 100049China

It is challenging to model the performance of distributed graph *** formulation cannot easily capture the diversified factors and complex interactions in the *** learning methods require a large number of training samples to generate an accurate prediction ***,it is time-consuming to run the required graph computation tests to obtain the training *** this paper,we propose TransGPerf,a transfer learning based solution that can exploit prior knowledge from a source scenario and utilize a manageable amount of training data for modeling the performance of a target graph computation *** results show that our proposed method is capable of generating accurate models for a wide range of graph computation tasks on PowerGraph and *** outperforms transfer learning methods proposed for other applications in the literature.

关键词： performance modeling distributed graph computation deep learning transfer learning

来源：评论

学校读者我要写书评

暂无评论

Accelerate Bit Manipulation in XiangShan Processer Using RISC-V B Extension and Instruction Fusion

Accelerate Bit Manipulation in XiangShan Processer Using RIS...

引用

2022 International Conference on Cloud computing, Internet of Things, and computer Applications, CICA 2022

作者： Zhang, Fawang Tang, Dan Cai, Ye College of Computer Science and Software Engineering Shenzhen University Shenzhen China State Key Laboratory of Computer Architecture Institute of Computing Technology Beijing China

ISBN: (纸本)9781510656673

In the field of high-performance computing, some application scenarios make extensive use of bit manipulation. RISC-V foundation issues B extension to reduce the number of instructions during the static compilation. But this extension does not cover the more frequent operations fully. Instruction fusion technology can fuse multiple instructions into one instruction at the decode stage to dynamically reduce the number of instructions. Therefore, this technology can supplement RISC-V B extension uncovered instruction, enhancing the processor’s capabilities for bit manipulation. This paper implements the RISC-V B extension 1.0 version on the Xiangshan processor. For accelerating more bit manipulation, we profiled general benchmarks by the simulator, proposed ten new instructions fusion targets, and implemented a fusion decode unit. Finally, we use Coremark and SPEC CPU 2006 integer to evaluate the performance. The Coremark results show that dynamic instructions reduce by 18.14%, and the performance improves by 12.09%. © 2022 SPIE.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

Energy-Efficient and Reliable Federated Learning in Heterogeneous Mobile-Edge computing 19

Energy-Efficient and Reliable Federated Learning in Heteroge...

引用

2022 IEEE SmartWorld, 19th IEEE International Conference on Ubiquitous Intelligence and computing, 2022 IEEE International Conference on Autonomous and Trusted Vehicles Conference, 22nd IEEE International Conference on Scalable computing and Communications, 2022 IEEE International Conference on Digital Twin, 8th IEEE International Conference on Privacy computing and 2022 IEEE International Conference on Metaverse, SmartWorld/UIC/ATC/ScalCom/DigitalTwin/PriComp/Metaverse 2022

作者： Xu, Yan Li, Liying Cong, Peijin Zhou, Junlong Nanjing University of Science and Technology School of Computer Science and Engineering China Institute of Computing Technology Chinese Academy of Sciences State Key Laboratory of Computer Architecture China

ISBN: (纸本)9798350346558

As a decentralized machine learning paradigm, Federated Learning (FL) is an emerging technique to protect user data privacy in Mobile Edge computing (MEC). FL adopts the idea of distributed privacy computing, enabling the terminal devices to train machine learning models required by servers locally and upload model parameters to servers for aggregation. However, in the process of coupling FL with MEC systems, device heterogeneity, malicious use of poisoned data by users and limited system resources will seriously affect FL training accuracy, system reliability and energy efficiency. In this paper, a highly reliable and energy-efficient FL scheme is designed to solve these issues. Specifically, we first propose a calculation method for estimating the reliability of heterogeneous devices. Then, we design a highly reliable device selection scheme based on devices' reliability, computing power, and participation times. Finally, we utilize the slack time in FL training to dynamically adjust the voltage and frequency of devices for reducing the energy consumption of training. Experimental results show that our proposed scheme improves the accuracy by up to 36.45% on average and saves energy by up to 48.63% when compared to the baseline and benchmarking methods. © 2022 IEEE.

关键词： Reliability

来源：评论

学校读者我要写书评

暂无评论

Energy-Constrained Task Scheduling in Heterogeneous Distributed Systems 26

Energy-Constrained Task Scheduling in Heterogeneous Distribu...

引用

26th International Conference on computer Supported Cooperative Work in Design, CSCWD 2023

作者： Chen, Chen Zhu, Jie Huang, Haiping Gao, Yingmeng Nanjing University of Posts & Telecommunications Nanjing210003 China Institute of Computing Technology Academy of Sciences State Key Laboratory Chinese of Computer Architecture Beijing100864 China Jiangsu High Technology Research Key Laboratory for Wireless Sensor Networks Nanjing210003 China

ISBN: (纸本)9798350331684

The resource-constrained task scheduling problem has been one of the popular research topics in cloud computing systems. By employing the dynamic voltage and frequency scaling (DVFS) techniques, the task scheduling can be further constrained by energy consumption. The paper investigates the DAG task scheduling considering both the resource and energy constraints in heterogeneous distributed systems. The objective is to minimize the scheduling length. An energy-constrained task scheduling framework is employed, where tasks are initially scheduled according to their upward rank values. Then two heuristics are proposed to improve the initial solution, namely, the simulated annealing local search method and the frequency adjustment method. Experiments are conducted by testing a large number of instances with multiple parameter settings, and the results show that the proposed algorithms are effective and efficient. © 2023 IEEE.

关键词： Energy utilization

来源：评论

学校读者我要写书评

暂无评论

Dynamic Task Allocation for Heterogeneous Multi-Robot System under Communication Constraints 6

Dynamic Task Allocation for Heterogeneous Multi-Robot System...

引用

6th IEEE Information technology, Networking, Electronic and Automation Control Conference, ITNEC 2023

作者： Li, Jun Cai, Zhongxuan Li, Minglong Huang, Wanrong Zhang, Yongjun National University of Defense Technology Institute for Quantum Information State Key Laboratory of High Performance Computing College of Computer China

ISBN: (纸本)9781665460033

Multi-robot system has important application potential in disaster rescue and other dangerous scenarios, in which task allocation is the basis for multi-robot cooperation to complete tasks. Due to the limited ability of individual robots, many complex scenarios requires coordination of different types of robot, e.g. robots for search, communication, rescue and so on. At the same time, communication is often limited in a wide range of scenarios. This brings two challenges in the realistic environment: heterogeneous multi-robot cooperation problem and communication constraint problem. This paper proposes a task allocation method based on hierarchical auction and dynamic grouping. By introducing a dynamic grouping algorithm to the general allocation procedure, we simultaneously optimize the number of robots in each group and the communication distance between robots. Then the communication loss rate can be reduced to improve task execution efficiency. Simulations of a communication constrained search and rescue scenario are conducted to verify the proposed algorithm, which are compared with the vanilla auction algorithm and a hierarchical approach on static grouping. The results show that our method surpasses the baseline on both task achievement and communication performance. © 2023 IEEE.

关键词： Multipurpose robots

来源：评论

学校读者我要写书评

暂无评论

LayerCFL:an efcient federated learning with layer-wised clustering

引用

Cybersecurity 2025年第1期8卷 72-85页

作者： Jie Yuan Rui Qian Tingting Yuan Mingliang Sun Jirui Li Xiaoyong Li School of Cyberspace Security Beijing University of Posts and TelecommunicationsBeijingChina Key Laboratory of Trustworthy Distributed Computing and Service(BUPT) Ministry of EducationBeijingChina Institute of Computer Science Faculty of Mathematics and Computer ScienceUniversity of GoettingenGöettingenGermany School of Information Technology Henan University of Chinese MedicineZhengzhouHenanChina

Federated Learning(FL)sufers from the Non-IID problem in practice,which poses a challenge for efcient and accurate model *** address this challenge,prior research has introduced clustered FL(CFL),which involves clustering clients and training them *** its potential benefts,CFL can be computationally and communicationally expensive when the data distribution is unknown *** is because CFL involves the entire neural networks of involved clients in computing the clusters during training,which can become increasingly timeconsuming with large-sized *** tackle this issue,this paper proposes an efcient CFL approach called LayerCFL that employs a Layer-wised clustering *** LayerCFL,clients are clustered based on a limited number of layers of neural networks that are pre-selected using statistical and experimental *** experimental results demonstrate the efectiveness of LayerCFL in mitigating the impact of Non-IID data,improving the accuracy of clustering,and enhancing computational efciency.

关键词： Federated learning Clustered federated learning Non-IID Layer-wised clustering

来源：评论

学校读者我要写书评

暂无评论

PIOD: An Efficient Parallel Iterative Algorithm for Solving Over-determined Equations 29

PIOD: An Efficient Parallel Iterative Algorithm for Solving ...

引用

29th IEEE International Conference on Parallel and Distributed Systems, ICPADS 2023

作者： Tian, Min Wei, Zhenguo Li, Yanlong Du, Wei Xiao, Lei Xu, Chaoshuai China Shandong Fundamental Research Center for Computer Science Shandong Provincial Key Laboratory of Computer Networks China Jinan Institute of Supercomputing Technology Jinan Key Laboratory of High Performance Computing China

ISBN: (纸本)9798350330717

The solution of over-determined equations plays a very important role in fields such as data fitting, signal processing, and machine learning. It is of great significance in predicting natural phenomena, optimizing engineering design, and other fields. However, there is currently no efficient method to solve over-determined equations, either being not accurate enough or consuming a lot of time. In this article, we propose a parallel iterative method for solving over-determined equations, called the PIOD algorithm. By using a sub-convergence condition to terminate the iterative calculation, we have developed a task partitioning strategy for the algorithm and implemented parallelization of the solution of over-determined equations on a distributed memory system. Our proposed algorithm achieves an average speedup of 152 ×times compared to the open-source Eigen solver. Additionally, it also achieves a parallel efficiency of over 30%. © 2023 IEEE.

关键词： conjugate gradient method MPI over-determined equations parallel algorithms parallel iterative methods

来源：评论

学校读者我要写书评

暂无评论

Automatic Target Description File Generation

引用

Journal of computer Science & technology 2023年第6期38卷 1339-1355页

作者：耿洪娜吕方钟茗崔慧敏薛景玲冯晓兵 State Key Laboratory of Processors Institute of Computing TechnologyChinese Academy of SciencesBeijing 100190China University of Chinese Academy of Sciences Beijing 100049China School of Computer Science and Engineering University of New South WalesSydneyNSW 2052Australia

Agile hardware design is gaining increasing momentum and bringing new chips in larger quantities to the market ***,it also takes new challenges for compiler developers to retarget existing compilers to these new chips in shorter time than ever ***,retargeting a compiler backend,e.g.,an LLVM backend to a new target,requires compiler developers to write manually a set of target description files(totalling 10300+lines of code(LOC)for RISC-V in LLVM),which is error-prone and *** this paper,we introduce a new approach,Au-tomatic Target Description File Generation(ATG),which accelerates the generation of a compiler backend for a new tar-get by generating its target description files *** a new target,ATG proceeds in two ***,ATG synthesizes a small list of target-specific properties and a list of code-layout templates from the target description files of a set of existing targets with similar instruction set architectures(ISAs).Second,ATG requests compiler developers to fill in the information for each instruction in the new target in tabular form according to the list of target-specific properties syn-thesized and then generates its target description files automatically according to the list of code-layout templates *** first stage can often be reused by different new targets sharing similar *** evaluate ATG using nine RISC-V instruction sets drawn from a total of 1029 instructions in LLVM *** enables compiler developers to gen-erate compiler backends for these ISAs that emit the same assembly code as the existing compiler backends for RISC-V but with significantly less development effort(by specifying each instruction in terms of up to 61 target-specific properties only).

关键词： retargetability compiler target description target backend automatic generator

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：