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 Nanjing 210044 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 China University of Petroleum (East 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： Combinatorial auction distributed DNN inference dynamic scheduling edge-cloud networks resource pricing

来源：评论

学校读者我要写书评

暂无评论

Modeling coalition formation for repeated games using learning approaches

Modeling coalition formation for repeated games using learni...

引用

2010 International Conference on computer Application and System Modeling, ICCASM 2010

作者： Wang, Zhong-Cun Wang, Chong-Lun Department of Computer Science and Technology National Key Laboratory for Novel Software Technology Nanjing University Nanjing 210093 China

ISBN: (纸本)9781424472369

In this paper, we introduce the notion of "weight" to task's capability, and describe the use of case-based learning and reinforcement learning in a coalition formation model when games are repeated. Based on the the notion "weight" we introduce, a weight-based coalition formation algorithm is proposed, but this algorithm can't always generate good coalitions, to supplement this, an randomized weight-based coalition formation algorithm is introduced. However, deciding when to use which algorithm is not such an easy thing, so a notion of "degree of similarity" is defined, through learning, an optimal degree of similarity can be attained to solve the above problem. In a word, we handle the coalition formation problem in a more of machine learning and data driven perspective.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Modeling Geographically Correlated Failures to Assess Network Vulnerability

Modeling Geographically Correlated Failures to Assess Networ...

引用

作者： Wang, Xiaoliang Chen, Mingzhi Lu, Sanglu Department of Computer Science and Technology National Key Laboratory for Novel Software Technology Nanjing University Nanjing210023 China

Current communication networks are facing more and more threats from large-scale regional damages, such as natural disasters (e.g., earthquake or tornado) and physical attacks (e.g., electromagnetic pulse attack or dragging anchors). Recently, several region failure models have been proposed to evaluate the impact of such geographically correlated failures on communication networks. These works mainly adopt a kind of 'deterministic' models, where network components within the affected area would fail simultaneously. Such failure models simplify the analysis but may fail to reflect some important behaviors of attacks and thus cause significant over-or under-estimation of region failures. To emulate the impact of realistic catastrophe events such as earthquake and tornado, this paper introduces two probabilistic failure models: 1) a concentric circle model and 2) a line segment model. In the probabilistic models, both the location and effect of the damage are treated as random events. The failure may randomly incur on the entire network plane, and the failure probability of a device depends on many factors, e.g., link length and distance to the damage center. We develop an efficient grid partition-based scheme to estimate the network vulnerability. Based on the grid partition scheme, we further develop a sampling scheme to significantly reduce the computation cost. The probability model helps us more deeply understand the network behaviors under region failure and facilitates the design and maintenance of future highly survivable mission critical networks. © 1972-2012 IEEE.

关键词： Earthquakes

来源：评论

学校读者我要写书评

暂无评论

LEARNING EFFICIENT ONLINE 3D BIN PACKING ON PACKING CONFIGURATION TREES 10

LEARNING EFFICIENT ONLINE 3D BIN PACKING ON PACKING CONFIGUR...

引用

10th International Conference on Learning Representations, ICLR 2022

作者： Zhao, Hang Yu, Yang Xu, Kai School of Computer Science National University of Defense Technology China National Key Lab for Novel Software Technology Nanjing University China

Online 3D Bin Packing Problem (3D-BPP) has widespread applications in industrial automation and has aroused enthusiastic research interest recently. Existing methods usually solve the problem with limited resolution of spatial discretization, and/or cannot deal with complex practical constraints well. We propose to enhance the practical applicability of online 3D-BPP via learning on a novel hierarchical representation - packing configuration tree (PCT). PCT is a full-fledged description of the state and action space of bin packing which can support packing policy learning based on deep reinforcement learning (DRL). The size of the packing action space is proportional to the number of leaf nodes, i.e. candidate placements, making the DRL model easy to train and well-performing even with continuous solution space. During training, PCT expands based on heuristic rules, however, the DRL model learns a much more effective and robust packing policy than heuristic methods. Through extensive evaluation, we demonstrate that our method outperforms all existing online BPP methods and is versatile in terms of incorporating various practical constraints. © 2022 ICLR 2022 - 10th International Conference on Learning Representationss. All rights reserved.

关键词： Heuristic methods

来源：评论

学校读者我要写书评

暂无评论

Uplink Control Information Based Clustered Access and Discontinuous Reception Management for mMTC

引用

IEEE Communications Letters 2025年

作者： Wang, Shao-Di Wang, Chang-Long Wang, Hui-Ming Zhou, Feng Leung, Victor C. M. Xidian University Key Laboratory of Electronic Information Countermeasure and Simulation Technology of Ministry of Education Xi’an710071 China Xi’an Jiaotong University School of Information and Communication Engineering Ministry of Education Key Lab for Intelligent Networks and Network Security Xi’an710049 China Shenzhen MSU BIT University Artificial Intelligence Research Institute Shenzhen518172 China Shenzhen University College of Computer Science and Software Engineering Shenzhen518060 China The University of British Columbia Department of Electrical and Computer Engineering VancouverBCV6T 1Z4 Canada

Uplink control information (UCI) and discontinuous reception (DRX) play important roles for massive machine type communication (mMTC). Despite their standalone significance, a conspicuous gap exists in comprehensively exploring the synergy between UCI and DRX concerning both energy conservation and latency reduction for mMTC. In this letter, we propose a UCI-based clustered access and DRX management (UCADM) scheme that strikes a balance between these two critical performance metrics, reducing energy usage while preserving low latency in uplink-downlink interaction for mMTC. Specifically, we present a strictly feasible solution for the proposed UCADM scheme by using the spectral radius and the constraint projection, which can effectively solve DRX switching operations as indicated by scheduling information from UCI. Simulation results confirm the effectiveness of our scheme. © 1997-2012 IEEE.

关键词： Information management

来源：评论

学校读者我要写书评

暂无评论

Deep learning-based prediction of individualized Real-time FSH doses in GnRH agonist long protocols

引用

Journal of Translational Medicine 2025年第1期23卷 1-11页

作者： Kong, Na Xia, Yu Wang, Zhilong Zhang, Hui Duan, Liyan Zhu, Yingchun Huang, Chenyang Yan, Guijun Mei, Jie Li, Wujun Sun, Haixiang Nanjing Drum Tower Hospital Drum Tower Clinical Medical College Nanjing University of Chinese Medicine Jiangsu Nanjing 210008 China Center for Reproductive Medicine and Obstetrics and Gynecology Nanjing Drum Tower Hospital Affiliated Hospital of Medical School Nanjing University Nanjing China Center for Molecular Reproductive Medicine Nanjing University Nanjing 2100 08 China National Institute of Healthcare Data Science at Nanjing University Jiangsu Nanjing 210008 China National Key Laboratory for Novel Software Technology Department of Computer Science and Technology Nanjing University Nanjing China Center for Medical Big Data Nanjing Drum Tower Hospital Affiliated Drum Tower Hospital Medical School of Nanjing University Jiangsu Nanjing 210008 China State Key Laboratory of Reproductive Medicine and China International Joint Research Center On Environment and Human Health Center for Global Health School of Public Health Nanjing Medical University Nanjing 211166 China Department of Reproductive Medicine Center Nanjing Drum Tower Hospital State Key Laboratory of Pharmaceutical Biotechnology Nanjing University Nanjing 210023 China

Background: Individualizing follicle-stimulating hormone (FSH) dosing during controlled ovarian stimulation (COS) is critical for optimizing outcomes in assisted reproduction but remains difficult due to patient heterogeneity. Most existing models are limited to static predictions of initial doses and do not support real-time adjustments throughout stimulation. Methods: We developed a deep learning model that integrates cross-temporal and cross-feature encoding (CTFE) to predict personalized daily FSH doses in patients undergoing COS using the GnRH agonist long protocol. A total of 13,788 IVF/ICSI cycles conducted between January 2018 and December 2020 were retrospectively analyzed. Women with baseline antral follicle counts between 7 and 30 were included. Data were randomly divided into training (n = 6761), validation (n = 2898), and test (n = 4135) sets. The model encodes both static (e.g., age, BMI, basic hormone levels) and dynamic (e.g., follicle development, hormone trends during COS) variables across stimulation days. Final dose predictions were generated using a K-nearest neighbor algorithm applied to low-dimensional latent representations derived from the deep encoder layers. Results: The CTFE model achieved a dose classification accuracy of 0.737 (± 0.004) and a weighted F1-score of 0.732 (± 0.005) on the test set. On key stimulation days 1 and 5, the CTFE model significantly outperformed traditional LASSO regression models (F1-score: 0.832 vs 0.699 on day 1;0.817 vs 0.523 on day 5;p < 0.001). Prediction performance was maintained beyond day 13 using a sliding window mechanism, despite reduced data availability in longer stimulation cycles. Conclusions: This is the first study to apply a cross-temporal and cross-feature deep learning framework for daily, individualized FSH dose prediction across the full duration of COS. The model demonstrated superior performance over conventional approaches and offers a promising tool for standardizing COS management. Al

关键词： Artificial intelligence Controlled ovarian stimulation Deep learning FSH dosing Personalized medicine

来源：评论

学校读者我要写书评

暂无评论

Automated text data extraction based on unsupervised small sample learning 7

引用

7th International Conference on Intelligent Systems and Knowledge Engineering, ISKE 2012 with 1st International Conference on Cognitive Systems and Information Processing, CSIP 2012

作者： Liu, Yulong Shi, Shengsheng Yuan, Chunfeng Huang, Yihua National Key Laboratory for Novel Software Technology Department of Computer Science and Technology Nanjing University Nanjing210093 China

ISBN: (纸本)9783642378287

Most of Web information extraction systems work with the DOM tree–based structured extraction rules to extract data records from Web pages;however, some, data items of, or even whole, of these data records are often in a semi-structured or unstructured text form. Thus, we need to introduce text data extraction rules to further extract the fine-grained data elements from those coarsegrained text items or records. However, generating text data extraction rules is a challenging task in either manual or automated way. In this paper, we propose an unsupervised learning approach to automatically deducing text data extraction rules from a small sample of text records. First of all, to prepare for extraction rule template deduction, we propose an iterative center core multiple sequence alignment method to align text columns in sample text records. Then, we propose an information entropy model based on the statistical features of text columns to further identify each column as either a template column or a data column. From identified template and data columns, plus some additional processing, we can quickly deduce the template, that is, the text data extraction rule. Eventually, we can use the text data extraction rule to perform the automated text data extraction from test text records. This unsupervised learning approach does not need any manual labeling and enables automated generation of text data extraction rules and text data extraction process. It is the first study effort toward the unsupervised small sample learning approach for automated text data extraction rule generation. The experimental results show that our approach achieves high accuracy. © Springer-Verlag Berlin Heidelberg 2014.

关键词： Websites

来源：评论

学校读者我要写书评

暂无评论

Study on Communication Model Between Heterogeneous Operating Systems for a Wireless Sensor Network Middleware

引用

电子学报(英文版) 2007年第3期16卷 543-546页

作者： LI Wenfeng WANG Ruchuan SUN Lijuan Department of Computer Science and Technology Nanjing University of Posts and Telecommunications Nanjing 210003 China State Key Laboratory for Novel Software Technology Nanjing University Nanjing 210093 China

Focusing on the problem that two HOS (Heterogeneous operating systems) of wireless sensor network can't communicate with each other, and based on the comparative analysis on common-use communication models for the node operating system, we had brought forward a heterogeneous operating system communication model for a wireless sensor network. It consists of two tactics: scheduling tactic and processing tactic, which work together to finish the heterogeneous operating system communication. We introduced mobile-agent mechanism and first come first serve scheduling algorithm to co-operate task scheduling in the model. Experimental results and preliminary applications show that the model provides a good way for heterogeneous operating system communication.

关键词：通信模型无线传感器网络通信技术异质性

来源：评论

学校读者我要写书评

暂无评论

Modular Verification of SPARCv8 Code

引用

Journal of computer science & technology 2020年第6期35卷 1382-1405页

作者： Jun-Peng Zha Xin-Yu Feng Lei Qiao Department of Computer Science and Technology Nanjing UniversityNanjing 210023China State Key Laboratory for Novel Software Technology Nanjing UniversityNanjing 210023China Beijing Institute of Control Engineering Beijing 100080China

Inline assembly code is common in system software to interact with the underlying hardware platforms. The safety and correctness of the assembly code is crucial to guarantee the safety of the whole system. In this paper, we propose a practical Hoare-style program logic for verifying SPARC (Scalable Processor Architecture) assembly code. The logic supports modular reasoning about the main features of SPARCv8 ISA (instruction set architecture), including delayed control transfers, delayed writes to special registers, and register windows. It also supports relational reasoning for refinement verification. We have applied it to verify that there is a contextual refinement between a context switch routine in SPARCv8 and a switch primitive. The program logic and its soundness proof have been mechanized in Coq.

关键词： Scalable Processor Architecture Version 8(SPARCv8) assembly code verification context switch Coq refinement verification

来源：评论

学校读者我要写书评

暂无评论

Service-Oriented Resource Allocation in Clouds： Pursuing Flexibility and Efficiency

引用

Journal of computer science & technology 2015年第2期30卷 421-436页

作者：张胜钱柱中吴杰陆桑璐 State Key Laboratory for Novel Software Technology Nanjing University Nanjing 210023 China Department of Computer Science and Technology Nanjing University Nanjing 210023 China Department of Computer and Information Sciences Temple University Philadelphia PA 19122 U.S.A

The networking-oblivious resource reservation model in today＇s public clouds cannot guarantee the performance of tenants＇ applications. Virtual networks that capture both computing and networking resource requirements of tenants have been proposed as better interfaces between cloud providers and tenants. In this paper, we propose a novel virtual network model that could specify not only absolute and relative location requirements but also time-varying resource demands. Building on top of our model, we study how to efficiently and flexibly place multiple virtual networks in a cloud, and we propose two algorithms, MIPA and SAPA, which focus on optimizing resource utilization and providing flexible placement, respectively. The mixed integer programming based MIPA transforms the placement problem into the multi-commodity flow problem through augmenting the physical network with shadow nodes and links. The simulated annealing-based SAPA achieves resource utilization efficiency through opportunistically sharing physical resources among multiple resource demands. Besides, SAPA allows cloud providers to control the trade-offs between performance guarantee and resource utilization, and between allocation optimality and running time, and allows tenants to control the trade-off between application performance and placement cost. Extensive simulation results confirm the efficiency of MIPA in resource utilization and the flexibility of SAPA in controlling trade-offs.

关键词： resource allocation virtual network embedding opportunistic resource sharing

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：