检索结果-内蒙古大学图书馆

3rd IEEE international workshop on System-on-Chip for Real-Time applications

作者： Halpin, B Sehgal, N Chen, CYR Syracuse Univ Intel Corp Santa Clara CA 95054 USA

ISBN: (纸本)076951944X

Increasing demands created by Systems-On-Chip (SOC) and process advances have increased the difficulty of timing driven placement. The primary issue in SOC is timing closure. This requires us to look at timing at all design levels, especially placement. Recently, several promising approaches for timing-driven placement have been presented using net length constraints for timing optimization [1, 2]. A Net Length Constraint (NLC) is an upper limit on a net's length. These net-constrained global placement techniques give excellent timing results by meeting NLCs on timing-critical nets. These works focused only on global NLC placement. Detailed placement and legalization are important steps in the placement flow. Current algorithms, which are not NLC aware, give back the gains from global NLC placement. The contributions of this paper are a new NLC global placement rebalancing method and two detailed placement algorithms that work in conjunction with the recursive bisection net-constrained global placer [1]. The first detailed placer uses grid-based placement and transportation solving to assign instances to the grid. The second detailed placer uses simulated annealing to optimize placement for NLC. On benchmark circuits from MCNC and Intel Corporation, the grid and simulated annealing placers are able to achieve placements which exceed constraints by, on average, only 2.7% and 1.9%, respectively.

关键词： Automatic control Circuit simulation Constraint optimization Delay Design engineering optimization methods Simulated annealing Timing Transportation Very large scale integration

来源：评论

学校读者我要写书评

暂无评论

An ASGC-Surrogate Model Based Yield optimization for Passive Microwave Devices with Differential Evolution Algorithm

An ASGC-Surrogate Model Based Yield Optimization for Passive...

引用

IEEE MTT-S international Microwave workshop Series on Advanced Materials and Processes for RF and THz applications (IMWS-AMP)

作者： Han, Yujie Li, Ping Shanghai Jiao Tong Univ Minist Educ China Res Design & Electromagnet Comp Key Lab Shanghai Peoples R China

ISBN: (纸本)9781665478342

The performance of microwave devices is affected by many uncertainties during the manufacturing process. Proper design with least influence from these disturbances is required for optimal yield. At the design stage of microwave devices, the electromagnetic (EM) based simulation methods are accurate and comprehensive, but it suffers from expensive computational cost. For yield optimization, highly efficient optimization algorithms are necessary. In this article, a global surrogate model is constructed on the basis of the adaptive hierarchical sparse grid collocation (ASGC) algorithm to significantly reduce the number of sampling points. The differential evolution algorithm combined with the ASGC-based surrogate model is proposed to decrease the total simulation time. Finally, to validate the accuracy and efficiency of the proposed algorithm, several representative examples are presented.

关键词： Adaptive hierarchical sparse grid collocation (ASGC) algorithm differential evolution (DE) algorithm microwave filters manufacturing uncertainties yield optimization

来源：评论

学校读者我要写书评

暂无评论

Hardware software partitioning using particle swarm optimization technique

Hardware software partitioning using particle swarm optimiza...

引用

6th international workshop on System-on-Chip for Real-Time applications

作者： Abdelhalim, M. B. Salama, A. E. Habib, S. E. -D. Cairo Univ Fac Engn Giza 12211 Egypt

ISBN: (纸本)9781424408986

In this paper we investigate the application of the Particle Swarm optimization (PSO) technique for solving the Hardware/Software partitioning problem. The PSO is attractive for the Hardware/Software partitioning problem as it offers reasonable coverage of the design space together with O(n) main loop's execution time, where n is the number of proposed solutions that will evolve to provide the final solution. We carried out several tests on a hypothetical, relatively-large Hardware/Software partitioning problem using the PSO algorithm as well as the Genetic Algorithm (GA), which is another evolutionary technique. We found that PSO outperforms GA in the cost function and the execution time. For the case of unconstrained design problem, we tested several hybrid combinations of PSO and GA algorithms;including PSO then GA, GA then PSO, GA followed by GA, and finally PSO followed by PSO. We found that a PSO followed by GA algorithm gives small or no improvement at all, while a GA then PSO algorithm gives the same results as the PSO alone. The PSO algorithm followed by another PSO round gave the best result as it allows another round of domain exploration. The second PSO round assign new randomized velocities to the particles, while keeping best particle positions obtained in the first round. We propose to name this successive PSO algorithm as the Re-excited PSO algorithm.

关键词： embedded systems hardware/software co-design hardware/software partitioning particle swarm optimization algorithm genetic algorithm evolutionary algorithms re-excited PSO

来源：评论

学校读者我要写书评

暂无评论

Prediction and optimization of Electric Resistance of Printed Graphene Nanosheet Based on Artificial Intelligence

Prediction and Optimization of Electric Resistance of Printe...

引用

IEEE MTT-S international Microwave workshop Series on Advanced Materials and Processes for RF and THz applications (IMWS-AMP)

作者： Chen, Hao Huang, Tian-Yu Zhang, Yi-Yang Liu, En-Cheng Yang, Wu Lu, Wei-Bing Southeast Univ Sch Informat Sci & Engn State Key Lab Millimeter Waves Nanjing Peoples R China Southeast Univ Chien Shiung Wu Coll Nanjing Peoples R China

ISBN: (纸本)9781665478342

Graphene has attracted wide attention in the field of electromagnetism and has become a front runner in fabricating electronic devices, especially flexible ones. However, currently, there's no report on the prediction of its electric resistance, which is critical to an electric component. In this work, several machine learning (ML) and artificial neural network (ANN) models are trained to predict the electric resistance(R) of printed graphene nanosheets with only the knowledge of the concentration of graphene ink, the width, and the length of the printed nanosheet. Three different ML algorithms, i.e., Kriging, Efficient Global optimization (EGO), and Support Vector regression (SVR) are exploited, together with ANNs with different sets of parameters. Harboring high accuracy and yielding satisfactory results, the models can be applied to predict the resistance of graphene nanosheets in real-time working conditions. It paves the way for the wide application of printable graphene nanosheets in electric devices.

关键词： Graphene resistance machine learning artificial neural network

来源：评论

学校读者我要写书评

暂无评论

Cellular Network optimization by Deep Reinforcement Learning and AI-Enhanced Ray Tracing 2

Cellular Network Optimization by Deep Reinforcement Learning...

引用

2nd ACM SIGSPATIAL international workshop on Spatial Big Data and AI for Industrial applications, GeoIndustry 2023

作者： Wang, Zhangyu Isci, Serkan Kanza, Yaron Kounev, Velin Shaqalle, Yusef University of California Santa Barbara Santa BarbaraCA United States At&t Labs - Research BedminsterNJ United States

ISBN: (纸本)9798400703508

In this paper we study the use of deep reinforcement learning that is supported by a ray tracer, on top of a detailed 3D model of the geospatial environment, for optimization of antenna tilts in cellular networks. We propose two novel mechanisms - -geospatial importance sampling and multi-path coefficient - -to efficiently pass geospatial information to the reinforcement learning model. We show that this approach can be used for fast and scalable optimization of tilt levels of cellular antennas. We present an experimental evaluation that compares the use of reinforcement learning to greedy search, simulated annealing and Bayesian optimization. Our study shows that reinforcement learning is effective and can cope with optimization problems that are at a greater scale than the settings the other algorithms can cope with. © 2023 Owner/Author.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Learning Surrogate Models of Document Image Quality Metrics for Automated Document Image Processing 13

Learning Surrogate Models of Document Image Quality Metrics ...

引用

13th IAPR international workshop on Document Analysis Systems (DAS)

作者： Singh, Prashant Vats, Ekta Hast, Anders Uppsala Univ Dept Informat Technol SE-75105 Uppsala Sweden

ISBN: (纸本)9781538633465

Computation of document image quality metrics often depends upon the availability of a ground truth image corresponding to the document. This limits the applicability of quality metrics in applications such as hyperparameter optimization of image processing algorithms that operate on-the-fly on unseen documents. This work proposes the use of surrogate models to learn the behavior of a given document quality metric on existing datasets where ground truth images are available. The trained surrogate model can later be used to predict the metric value on previously unseen document images without requiring access to ground truth images. The surrogate model is empirically evaluated on the Document Image Binarization Competition (DIBCO) and the Handwritten Document Image Binarization Competition (H-DIBCO) datasets.

关键词： surrogate models document image quality metrics hyperparameter optimization

来源：评论

学校读者我要写书评

暂无评论

Alternative decompositions for distributed maximization of network utility: Framework and applications

Alternative decompositions for distributed maximization of n...

引用

IEEE INFOCOM 2006 Conference/25th IEEE international Conference on Computer Communications

作者： Palomar, Daniel P. Chiang, Mung Princeton Univ Dept Elect Engn Princeton NJ 08544 USA

ISBN: (纸本)9781424402212

Network utility maximization (NUM) problems provide an important approach to conduct network resource management and to view layering as optimization decomposition. In the existing literature, distributed implementations are typically achieved by the means of the so-called dual decomposition technique. However, the span of decomposition possibilities includes many other elements which thus far have not been fully exploited, such as the use of the primal decomposition technique, the versatile introduction of auxiliary variables, and the potential of multilevel decompositions. This paper presents a systematic framework to exploit the potential of the alternative decomposition structures as a way to obtain different distributed algorithms, each with a different tradeoff among convergence speed, message passing amount and asymmetry, and distributed computation architecture. Many specific applications are considered to illustrate the proposed framework, including resource-constrained and direct-control rate allocation, and rate allocation among QoS classes and with multipath routing. For each of these applications, the associated generalized NUM formulation is first presented, followed by the development of novel alternative decompositions and numerical experiments on the resulting new distributed algorithms.

关键词： rate control congestion control resource allocation mathematical programming/optimization network utility maxinlization distributed algorithm network control by pricing

来源：评论

学校读者我要写书评

暂无评论

Proceedings of the 5th workshop on Mapping of applications to MPSoCs and 15th international workshop on Software and Compilers for Embedded Systems, Map2MPSoC/SCOPES 2012

Proceedings of the 5th Workshop on Mapping of Applications t...

引用

5th workshop on Mapping of applications to MPSoCs, Map2MPSoC 2012 and 15th international workshop on Software and Compilers for Embedded Systems, SCOPES 2012

ISBN: (纸本)9781450313360

The proceedings contain 7 papers. The topics discussed include: CLASSY: a clock analysis system for rapid prototyping of embedded applications on MPSoCs;decoupled inter- and intra- application scheduling for composable and robust embedded SoC platforms;exploration of heuristic scheduling algorithms for 3D multicore processors;hardware virtualization-driven software task switching in recon gurable multi-processor system-on-chip architectures;debugging component-based embedded applications;fast architecture evaluation of heterogeneous MPSoCs by host-compiled simulation;and dynamic method to evaluate code optimization effectiveness.

关键词：

来源：评论

学校读者我要写书评

暂无评论

STABILIZED BARZILAI-BORWEIN METHOD

引用

Journal of Computational Mathematics 2019年第6期37卷 916-936页

作者： Oleg Burdakov Yuhong Dai Na Huang Department of Mathematics Linkoping UniversityLinkopingSweden LSEC ICMSECAcademy of Mathematics and Systems ScienceChinese Academy of SciencesBeijing 100190China Department of Applied Mathematics College of ScienceChina Agricultural UniversityBeijing 100083China

The Barzilai-Borwein(BB)method is a popular and efficient tool for solving large-scale unconstrained optimization *** search direction is the same as for the steepest descent(Cauchy)method,but its stepsize rule is *** to this,it converges much faster than the Cauchy method.A feature of the BB method is that it may generate too long steps,which throw the iterates too far away from the ***,it may not converge,even when the objective function is strongly *** this paper,a stabilization technique is *** consists in bounding the distance between each pair of successive iterates,which often allows for decreasing the number of BB *** the BB met hod does not converge,our simple modification of this method makes it *** strongly convex functions with Lipsch计s gradients,we prove its global convergence,despite the fact that no line search is involved,and only gradient values are *** the number of stabilization steps is proved to be finite,the stabilized version inherits the fast local convergence of the BB met *** presented results of extensive numerical experiments show that our stabilization technique often allows the BB method to solve problems in a fewer计erations,or even to solve problems where the latter fails.

关键词： Unconstrained optimization Spectral algorithms Stabilization Convergence analysis.

来源：评论

学校读者我要写书评

暂无评论

Applying Intelligent optimization algorithms for Evaluation of Transformer Black Box Model 6th

Applying Intelligent Optimization Algorithms for Evaluation ...

引用

6th international workshop on Soft Computing applications (SOFA)

作者： Bigdeli, Mehdi Valii, Mohammad Azizian, Davood Islamic Azad Univ Zanjan Branch Dept Elect Engn Zanjan Iran Islamic Azad Univ Abhar Branch Dept Elect Engn Abhar Iran

ISBN: (纸本)9783319184166;9783319184159

To study the behavior of the transformers and the effects of these devices on power network performance in transient states, various models with different objectives are presented. One of the most important objectives is to investigate the overvoltages made in transformer as a result of lightning waves. For this reason, a black box model applied in the analysis of transient state of the transformer high frequency (up to 1.2 MHz). As a new work, the Genetic Algorithm (GA), Particle Swarm optimization (PSO), and Imperialist Competitive Algorithm (ICA) are proposed in the estimation of parameters of black box model using the necessary measurements on three-phase transformer 2.5 MVA and 6300/420 V and then compared by analytical methods.

关键词： Transformer Transient analysis Black box model Intelligent algorithms

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：