检索结果-内蒙古大学图书馆

41st International Conference on Machine Learning, ICML 2024

作者： Wang, Junfu Guo, Yuanfang Yang, Liang Wang, Yunhong State Key Laboratory of Software Development Environment Beihang University Beijing China School of Computer Science and Engineering Beihang University Beijing China Shen Yuan Honors College Beihang University Beijing China School of Artificial Intelligence Hebei University of Technology Tianjin China

Graphs with heterophily have been regarded as challenging scenarios for Graph Neural Networks (GNNs), where nodes are connected with dissimilar neighbors through various patterns. In this paper, we present theoretical understandings of heterophily for GNNs by incorporating the graph convolution (GC) operations into fully connected networks via the proposed Heterophilous Stochastic Block Models (HSBM), a general random graph model that can accommodate diverse heterophily patterns. Our theoretical investigation comprehensively analyze the impact of heterophily from three critical aspects. Firstly, for the impact of different heterophily patterns, we show that the separability gains are determined by two factors, i.e., the Euclidean distance of the neighborhood distributions and pE [deg], where E [deg] is the averaged node degree. Secondly, we show that the neighborhood inconsistency has a detrimental impact on separability, which is similar to degrading E [deg] by a specific factor. Finally, for the impact of stacking multiple layers, we show that the separability gains are determined by the normalized distance of the lpowered neighborhood distributions, indicating that nodes still possess separability in various regimes, even when over-smoothing occurs. Extensive experiments on both synthetic and real-world data verify the effectiveness of our theory. Copyright 2024 by the author(s)

关键词：

来源：评论

学校读者我要写书评

暂无评论

Probabilistic Shaping for Rotationally Symmetrical Two-Dimensional Constellations 35

Probabilistic Shaping for Rotationally Symmetrical Two-Dimen...

引用

35th IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC 2024

作者： Yuan, Ruimin Yao, Xinyuanmeng Yang, Jiayi Bai, Baoming Ma, Xiao Xidian University State Key Lab. of Isn Xi'an710071 China Sun Yat-sen University School of Computer Science and Engineering Guangzhou510006 China Sun Yat-sen University School of Electronics and Communication Engineering Shenzhen518107 China

ISBN: (纸本)9798350362244

A ring-based constant composition distribution matcher (R-CCDM) is designed for shaping rotationally symmetrical two-dimensional (RS-2D) constellations. The procedure of distribution matching can be divided into two steps. First, nonuniformly distributed rings are generated by CCDM to approach a probability distribution calculated from that of modulation symbols. Second, a uniform mapping is performed to select one amplitude signal point from each ring. Compared with CCDM, the proposed DM contains multiple amplitude compositions, and hence achieves less rate loss. Numerical results show that, at moderate spectral efficiency, probabilistic shaping with R-CCDM outperforms that with CCDM by about 0.2 ∼ 0.3 ~dB for different modulation formats. © 2024 IEEE.

关键词： Frequency division multiplexing

来源：评论

学校读者我要写书评

暂无评论

A Service-Enhanced Task Offloading Method in MEC-Enabled IoV Networks 19

A Service-Enhanced Task Offloading Method in MEC-Enabled IoV...

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： Zhao, Bohai Peng, Kai Leung, Victor C.M. Xia, Yunni Huaqiao University College of Engineering Quanzhou China Nanjing University State Key Lab. for Novel Software Technology Nanjing China Shenzhen University School of Computer Science and Software Engineering Shenzhen China The University of British Columbia Department of Electrical and Computer Engineering Vancouver Canada Chongqing University School of Computers Chongqing China

ISBN: (纸本)9798350346558

To facilitate the evolution from independent transportation means to vehicle-to-infrastructure communications, traditional vehicle-to-vehicle (V2V) technologies are growing to a more advanced and generalized paradigm, i.e., the Internet of Vehicles (IoV). This paradigm has stimulated ever-increasing calling for more capabilities and intelligence for vehicles to foster the potential of supporting a plethora of emerging applications and ubiquitous automotive network access. Combining caching-aided communication and edge computing has been expected to be an advantageous framework to cope with the surging requirements for timely data processing and storage from intelligent IoV applications. In light of the delay restriction and the heterogeneity of edge service nodes (ESNs), we study the service-enhanced problem in edge-enabled IoV networks by configuring proactive service caching. Specifically, based on task scale and relational restrictions, we design an adaptive non-linear caching probability mapping function and further propose a proactive service-enhanced task offloading method in edge-enabled IoV networks named SEOEV. Performance evaluation demonstrates that SEOEV efficiently reduces task processing overhead in comparison to other comparative computing migration approaches. © 2022 IEEE.

关键词： Multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

dgQuEST: Accelerating Large Scale Quantum Circuit Simulation through Hybrid CPU-GPU Memory Hierarchies 18th

dgQuEST: Accelerating Large Scale Quantum Circuit Simulation...

引用

18th IFIP WG 10.3 International Conference on Network and Parallel Computing, NPC 2021

作者： Feng, Tianyu Chen, Siyan You, Xin Zhong, Shuzhang Yang, Hailong Luan, Zhongzhi Qian, Depei State Key Laboratory of Software Development Environment Beihang University Beijing China School of Computer Science and Engineering Beihang University Beijing China

ISBN: (纸本)9783030935702

With the advancement of quantum computing, verifying the correctness of the quantum circuits becomes critical while developing new quantum algorithms. Constrained by the obstacles of building practical quantum computers, quantum circuit simulation has become a feasible approach to develop and verify quantum algorithms. Although there are many quantum simulators available, they either achieve low performance on CPUs, or limited simulation scale (e.g., number of qubits) on GPUs due to limited memory capacity. Therefore, we propose dgQuEST, a novel acceleration method that utilizes hybrid CPU-GPU memory hierarchies for large-scale quantum circuit simulation across multiple nodes. dgQuEST adopts efficient memory management and communication schemes to leverage the distributed CPU and GPU memories for accelerating large-scale quantum simulation. Our evaluation demonstrates that dgQuEST achieves an average speedup of 403 × compared to QuEST on quantum circuit simulation with 32 qubits, and scales to quantum circuit simulation with 35 qubits on two GPU nodes, far beyond the state-of-the-art implementation HyQuas can support. © 2022, IFIP International Federation for Information Processing.

关键词： Circuit simulation

来源：评论

学校读者我要写书评

暂无评论

Utilization-Aware Data Variable Allocation on NVM- Based SPM in Real-Time Embedded Systems

引用

Journal of Electronic Science and Technology 2021年第2期19卷 163-172页

作者： Jin-Yu Zhan Yi-Xin Li Wei Jiang Jun-Huan Yang School of Information and Software Engineering University of Electronic Science and Technology of ChinaChengdu 610054 State Key Lab of Mathematical Engineering and Advanced Computing Wuxi 214000

With the development of the nonvolatile memory(NVM),using NVM in the design of the cache and scratchpad memory(SPM)has been *** paper presents a data variable allocation(DVA)algorithm based on the genetic algorithm for NVM-based SPM to prolong the *** lifetime can be formulated indirectly as the write counts on each SPM *** the differences between global variables and stack variables,our optimization model has three *** constraints of the central processing unit(CPU)utilization and size are used for all variables,while no-overlay constraint is only used for stack *** satisfy the constraints of the optimization model,we use the greedy strategy to generate the initial population which can determine whether data variables are allocated to SPM and distribute them evenly on SPM ***,we use the Mälardalen worst case executive time(WCET)benchmark to evaluate our *** experimental results show that the DVA algorithm can not only obtain close-to-optimal solutions,but also prolong the lifetime by 9.17% on average compared with SRAM-based SPM.

关键词： Data variables allocation genetic algorithm nonvolatile momory(NVM) scratchpad memory(SPM)

来源：评论

学校读者我要写书评

暂无评论

DRStencil: Exploiting Data Reuse within Low-order Stencil on GPU 23

DRStencil: Exploiting Data Reuse within Low-order Stencil on...

引用

23rd IEEE International Conference on High Performance Computing and Communications, 7th IEEE International Conference on Data Science and Systems, 19th IEEE International Conference on Smart City and 7th IEEE International Conference on Dependability in Sensor, Cloud and Big Data Systems and Applications, HPCC-DSS-SmartCity-DependSys 2021

作者： You, Xin Yang, Hailong Jiang, Zhonghui Luan, Zhongzhi Qian, Depei State Key Laboratory of Software Development Environment Beijing100191 China School of Computer Science and Engineering Beihang University Beijing100191 China

ISBN: (纸本)9781665494571

Stencil computation is widely adopted in scientific applications as one of the most significant computation patterns. Although there are various optimizations proposed to accelerate the stencil computation, the low-order stencil still suffers from limited performance on GPU due to its low computation inten-sity. In this paper, we propose the fusion-partition optimization techniques to accelerate the low-order stencil computation and implement an effective code generation framework DRStencil to automatically generate optimized stencil codes with fusion-partition applied. Specifically, we adopt a four-stage optimization workflow such as time-fusion, partition, forward and backward computation. We also propose an auto-tuning method to deter-mine the optimal parameter settings of the generated stencil codes. We evaluate DRStencil with representative low-order stencils on Nvidia P100, V100, and A100 GPUs. Our evaluation results achieve 1.46 x, 1.59 x, and 1.10 x speedup on average for widely used low-order stencils compared to the state-of-the-art implementations on P100, V100, and A100 GPUs, respectively. © 2021 IEEE.

关键词： Codes Smart cities Graphics processing units Manuals Parallel processing Computational efficiency Optimization

来源：评论

学校读者我要写书评

暂无评论

Diving into Continual Ultra-fine-grained Visual Categorization

Diving into Continual Ultra-fine-grained Visual Categorizati...

引用

2023 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2023

作者： Zhang, Pengcheng Yu, Xiaohan Bai, Xiao Zheng, Jin Wu, Xiaoyu Gao, Yongsheng Beihang University School of Computer Science and Engineering State Key Laboratory of Software Development Environment Jiangxi Research Institute Beijing China Griffith University Institute for Integrated and Intelligent Systems Nathan4111 Australia

ISBN: (纸本)9798350382204

Recent advance in ultra-fine-grained visual categorization (ultra-FGVC) has significantly boosted the capability of deep neural networks for ultra-FGVC tasks. However, building models for continually learning to recognize increasing ultra-fine-grained categories is still under-explored. This limits the application of ultra-FGVC techniques in real-world production. To this, we take the first attempt for continual ultra-FGVC. By evaluating existing continual learning methods on the constructed continual ultra-FGVC benchmark, we observe that the main bottleneck lies in the limited model plasticity for incrementally adapting to new tasks. This can be caused by excessive anti-forgetting constraints as the difficult ultra-FGVC task requires substantial update of parameters, and over-fitting on early tasks given that the ultra-fine-grained categories are with very few training samples. To tackle these problems, we propose a joint self-supervised learning and prompting model. The prompt-based continual learning framework offers proper anti-forgetting operation by fixed pretrained vision transformer and adaptive prompt selection. By jointly optimizing the learnable prompts with an adversarial self-supervised loss, the over-fitting on each continual learning task is mitigated. Extensive experiments demonstrate that the proposed method outperforms existing continual learning methods on the challenging continual ultra-FGVC problem. © 2023 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

MiTAR:a study on human activity recognition based on NLP with microscopic perspective

引用

Frontiers of computer Science 2021年第5期15卷 93-110页

作者： Huichao MEN Botao WANG Gang WU School of Computer Science and Engineering Northeastern UniversityShenyang 110000China State Key Laboratory for Novel Software Technology Nanjing UniversityNanjing 210000China

Nowadays,human activity recognition is becoming a more and more significant topic,and there is also a wide range of applications for it in real world *** data is an important data source in engineering and *** present,some studies have been carried out in the field of human activity recognition based on sensor data in a macroscopic ***,many studies in this perspective face some *** pivotal limitation is uncontrollable data segment length of different kinds of *** feature and data form are also influencing *** paper carries out the study creatively on a microscopic perspective with an emphasis on the logic and relevance between data segments,attempting to apply the idea of natural language processing and the method of data symbolization to the study of human activity recognition and try to solve the problem *** this paper,several activity-element definitions and three algorithms are proposed,including the algorithm of dictionary building,the algorithm of corpus building,and activity recognition algorithm improved from a natural language analysis method,*** experiments on different aspects of this model are *** experiments are carried out on six complex and representative single-level sensor datasets,namely UCI Sports and Daily dataset,Skoda dataset,WISDM Phoneacc dataset,WISDM Watchacc dataset,Healthy Older People dataset and HAPT dataset,which prove that this model can be applied to different datasets and achieve a satisfactory recognition result.

关键词： human activity recognition NLP time series data analysis data symbolization

来源：评论

学校读者我要写书评

暂无评论

Delving into the Reversal Curse: How Far Can Large Language Models Generalize? 38

Delving into the Reversal Curse: How Far Can Large Language ...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Lin, Zhengkai Fu, Zhihang Liu, Kai Xie, Liang Lin, Binbin Wang, Wenxiao Cai, Deng Wu, Yue Ye, Jieping State Key Lab of CAD&CG Zhejiang University China Alibaba Cloud China College of Biomedical Engineering & Instrument Science Zhejiang University China College of Computer Science and Technology Zhejiang University of Technology China School of Software Technology Zhejiang University China Fullong Inc. China

While large language models (LLMs) showcase unprecedented capabilities, they also exhibit certain inherent limitations when facing seemingly trivial tasks. A prime example is the recently debated "reversal curse", which surfaces when models, having been trained on the fact "A is B", struggle to generalize this knowledge to infer that "B is A". In this paper, we examine the manifestation of the reversal curse across various tasks and delve into both the generalization abilities and the problem-solving mechanisms of LLMs. This investigation leads to a series of significant insights: (1) LLMs are able to generalize to "B is A" when both A and B are presented in the context as in the case of a multiple-choice question. (2) This generalization ability is highly correlated to the structure of the fact "A is B" in the training documents. For example, this generalization only applies to biographies structured in "[Name] is [Description]" but not to "[Description] is [Name]". (3) We propose and verify the hypothesis that LLMs possess an inherent bias in fact recalling during knowledge application, which explains and underscores the importance of the document structure to successful learning. (4) The negative impact of this bias on the downstream performance of LLMs can hardly be mitigated through training alone. These findings offer a novel perspective on interpreting LLMs' generalization through their intrinsic mechanisms and provide insights for developing more effective learning methods. © 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

A Case Study of Dependency Network for Building Packages: The Fedora Linux Distribution 35

A Case Study of Dependency Network for Building Packages: Th...

引用

35th International Conference on software engineering and Knowledge engineering, SEKE 2023

作者： Du, Jiman Zhu, Jiaxin Li, Hui Chen, Wei Xu, Lijie Liu, Jie Chen, Zhifeng School of Computer Electronics and Information Guangxi University China State Key Lab of Computer Science ISCAS University of CAS China University of Chinese Academy of Sciences Nanjing China Nanjing Institute of Software Technology China MIIT Key Lab of Cloud Computing Standards and Applications China Electronic Standardization Institute China

To port the Linux distributions to a new Instruction Set Architecture (ISA), developers have to rebuild the software packages of the distributions. The complex dependencies of the software packages bring a great challenge. It is important to understand and properly handle the dependencies. We selected Fedora, a typical Linux distribution, and studied the dependencies within the software repositories of aarch64 and x86_64 architecture. We proposed a package dependency network framework to study the roles played by different packages. We obtained three network dependency patterns and proposed the corresponding division strategies which help developer build the source packages in parallel. Our study reveals that the key packages located at the root of multiple dependency chains significantly impact the division of the network, and their builds should be prioritized. Meanwhile, some packages with external dependencies can be temporarily masked to make a sub-network independent. Furthermore, the network dependency patterns are also observed in Fedora 33 riscv64 and OpenEuler riscv64. Our findings can help researchers have a better knowledge of Linux distribution dependency network and help practitioners conduct efficient package builds. © 2023 Knowledge Systems Institute Graduate school. All rights reserved.

关键词： software packages

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：