检索结果-内蒙古大学图书馆

International conference on Information Engineering and Computer Science

作者： Yin, Qiang Qi, Yibo Xiao, Jiaqing Department of Mathematics School of Science Wuhan University of Technology Wuhan 430070 China

ISBN: (纸本)9781424479412

Though many numerical methods have been put for nonlinear equations, their convergence and performance are highly sensitive to the initial guesses of the solution pre-supplied. However, the selection of good initial guess is often of hard work. Aiming at this, a novel approach is proposed to resolve nonlinear equations. It takes genetic algorithms' new achievement differential evolution algorithms as the main technique. With a function deflection technique and a novel space contraction method to reinitialize, it resolve nonlinear equations by transform them into correspondent optimization problems. Convergence reliability, computational cost and applicability of different algorithms were compared by testing several classical nonlinear equations and a benchmark mechanics problem. The numerical experiments done show that the put approach has reliable convergence probability, high convergence rate and solution precision. And DE is a successful approach in solving equations both in theory and application. ©2010 IEEE.

关键词： nonlinear equations

来源：评论

学校读者我要写书评

暂无评论

A mode-locked fiber laser simulation platform for self-tuning algorithm optimization 11

A mode-locked fiber laser simulation platform for self-tunin...

引用

conference on Optoelectronic Devices and Integration XI held as Part of SPIE/COS Photonics ASIA conference

作者： Ma, Qiuying Yu, Haoyang Zhou, Qian Li, Xinghui Wu, Guanhao Ni, Kai Tsinghua Univ Tsinghua Shenzhen Int Grad Sch Div Adv Mfg Shenzhen 518055 Peoples R China Cent South Univ Dept Automat Changsha 410083 Peoples R China

ISBN: (纸本)9781510656956;9781510656949

Mode-locked fiber lasers based on nonlinear polarization rotation (NPR) have been widely applied due to the simple setup, high performance and rich nonlinear dynamics. However, temperature, vibration, and stress can easily disrupt the optimized mode-locked state. To address this problem, automatic mode-locked lasers using different self-tuning algorithms are proposed in recent years. However, it is relatively difficult to verify and optimize the performance of self-tuning algorithm since the use of actual laser platforms, which hinders the development of intelligent mode-locked fiber laser. In this paper, we demonstrate a simulation platform for NPR mode-locked fiber lasers by using coupled Ginzburg-Landau equation and Jones matrix, which makes the optimization of intelligent self-tuning algorithm easier. As a proof-of-principle demonstration, genetic algorithm and human-like algorithm are implemented to prove the ability of comparing different self-tuning algorithms.

关键词： Mode-locked laser NPR Ginzburg-Landau equation simulation platform

来源：评论

学校读者我要写书评

暂无评论

performance Optimisation and Productivity Centre of Excellence 14

Performance Optimisation and Productivity Centre of Excellen...

引用

14th International conference on high performance Computing & Simulation (HPCS)

作者： Bridgwater, Sally Numer Algorithms Grp Oxford England

ISBN: (纸本)9781509020881

As machines get larger and scientific applications advance, it is more and more imperative to fully utilize high performance computing (HPC) capability. The complexity and changing landscape of parallel computers may lead to users being unable or unsure how to achieve optimal performance from their applications and fully utilize their HPC resources. The performance Optimisation and Productivity Centre of Excellence in Computing Applications (POP) has received funding from the European Commission as part of the Horizon 2020 programme to help alleviate these issues. It aims to uncover inefficiencies and their causes in existing parallel HPC applications that will lead to an improvement in the productivity and competitiveness of European organizations, in academia, government and industry. The POP project will drive efforts to highlight the need for and best practices in performance optimization through performance audits on codes along with training events to improve knowledge in this area. The aim is to help developers target their code development and refactoring in the most efficient direction and provide a return on investment from the savings due to the performance improvement. The POP project combines the expertise and experience of Barcelona Supercomputing Center (BSC), high performance Computing Center Stuttgart, Jülich Supercomputing Centre (JSC), Numerical algorithms Group Ltd (NAG), RWTH Aachen and TERATEC. This combination provides longstanding and well respected resources in the academic and commercial realms. The POP members have come together to create a coherent and consistent methodology to give a clear, precise and useful overview of the performance of each HPC application. The services of the POP project are free of charge to organizations with in the EU to analyze and advise on any parallel code in academic, government or industrial organizations of any domain. © 2016 IEEE.

关键词： component profiling Paraver Scalasca Centre of Excellence HPC scalability performance optimization

来源：评论

学校读者我要写书评

暂无评论

Sequential Classification-Based optimization for Direct Policy Search 31

Sequential Classification-Based Optimization for Direct Poli...

引用

31st AAAI conference on Artificial Intelligence

作者： Hu, Yi-Qi Qian, Hong Yu, Yang Nanjing Univ Natl Key Lab Novel Software Technol Nanjing 210023 Jiangsu Peoples R China Collaborat Innovat Ctr Novel Software Technol & I Nanjing 210023 Jiangsu Peoples R China

Direct policy search often results in high-quality policies in complex reinforcement learning problems, which employs some optimization algorithms to search the parameters of the policy for maximizing the its total reward. Classification based optimization is a recently developed framework for derivative-free optimization, which has shown to be effective and efficient for non-convex optimization problems with many local optima, and may provide a power optimization tool for direct policy search. However, this framework requires to sample a batch of solutions for every update of the search model, while in reinforcement learning, the environment often offers only sequential policy evaluation. Thus the classification-based optimization may not efficient for direct policy search, where solutions have to be sampled sequentially. In this paper, we adapt the classification-based optimization for sequential sampled solutions by forming the sample batch via reusing historical solutions. Experiments on a helicopter hovering task and controlling tasks in OpenAI Gym show that the new algorithm significantly improve the performance from several state-of-the-art derivative-free optimization approaches.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

JOTA: Joint optimization for the task assignment of sketch-based measurement

引用

COMPUTER COMMUNICATIONS 2017年 102卷 17-27页

作者： Su, Zhiyang Wang, Ting Hamdi, Mounir Hong Kong Univ Sci & Technol Kowloon Hong Kong Peoples R China Hamad Bin Khalifa Univ Doha Qatar

Sketch-based measurement provides traffic data summary in a memory-efficient way with provable accuracy bound. Recent advances in software defined networking facilitate the development and implementation of sketch-based measurement applications. However, sketch-based measurement usually requires TCAMs and SRAMs which are precious resource in switch to match packet fields. The key, challenge for sketch-based measurement is how to accept more concurrent measurement tasks with minimum resource usage. Existing proposals attempt to achieve this goal by exploring different task assignment algorithms. We argue that by sacrificing a small amount of accuracy, the resource usage can be decreased dramatically. In this paper, we propose JOTA, a novel system to improve the performance of the task assignment for sketch-based measurement. By considering both TCAM and SRAM capacities, we formulate the problem as a mixed integer nonlinear programming problem. Due to its high computational complexity, we divide the initial problem into two stages and develop a two-stage heuristic to efficiently produce task assignment. In particular, we present an algorithm which guarantees (1 + beta) approximation ratio to solve the second stage task assignment. Extensive experiments with three different measurement tasks and real packet traces demonstrate that JOTA significantly reduces the resource usage and accepts 33% more tasks compared with the first fit approach. (C) 2017 Elsevier B.V. All rights reserved.

关键词： software defined networking Network management Network measurement

来源：评论

学校读者我要写书评

暂无评论

high-Speed Turbo Equalization for GPP-based software Defined Radios

High-Speed Turbo Equalization for GPP-based Software Defined...

引用

IEEE Military Communications conference

作者： Schwall, Michael Jondral, Friedrich K. Karlsruhe Inst Technol D-76021 Karlsruhe Germany

ISBN: (纸本)9780769551241

high data rate waveforms for software defined radios (SDR) have to cope with frequency selective fading due to the mobile use in different harsh transmission environments. The received signal needs to be equalized in order to restore the transmitted information. Turbo equalization is a promising approach to deal with the inter-symbol interference occurring at the receiver. The iterative exchange of soft information between the equalizer and the decoder improves the decision reliability and hence reduces the bit error probability compared to conventional receivers. However, the necessary resource-demanding soft-input soft-output algorithms require a high processing performance to ensure real-time capability. In this paper, we will present a highspeed implementation of a turbo equalizer for SDRs with digital signal processing being performed on general purpose processors (GPP). The implementation will utilize linear MMSE filtering and suboptimal algorithms like overlapping sub-trellis MAX-Log-MAP decoding, approximations of mathematical operations, parallelization methods such as threading-based pipelining, and processor specific optimizations like single instruction multiple data (SIMD) commands. We will present the processing gains for each optimization level, highlight the performance loss for the suboptimal modifications and analyze the latency introduced by the pipelined processing. So far, transmissions with data rates up to 5.4 Mbit/s can be decoded in real-time with negligible performance loss and tolerable delay.

关键词： equalization gpp gpu high-speed implementation multi-core sdr turbo

来源：评论

学校读者我要写书评

暂无评论

Gradient compression for communication-limited convex optimization 57

Gradient compression for communication-limited convex optimi...

引用

57th IEEE conference on Decision and Control (CDC)

作者： Khirirat, Sarit Johansson, Mikael Alistarh, Dan Royal Inst Technol KTH Dept Automat Control Sch Elect Engn Stockholm Sweden Royal Inst Technol KTH ACCESS Linnaeus Ctr Stockholm Sweden IST Austria Vienna Austria

ISBN: (纸本)9781538613955

Data-rich applications in machine-learning and control have motivated an intense research on large-scale optimization. Novel algorithms have been proposed and shown to have optimal convergence rates in terms of iteration counts. However, their practical performance is severely degraded by the cost of exchanging high-dimensional gradient vectors between computing nodes. Several gradient compression heuristics have recently been proposed to reduce communications, but few theoretical results exist that quantify how they impact algorithm convergence. This paper establishes and strengthens the convergence guarantees for gradient descent under a family of gradient compression techniques. For convex optimization problems, we derive admissible step sizes and quantify both the number of iterations and the number of bits that need to be exchanged to reach a target accuracy. Finally, we validate the performance of different gradient compression techniques in simulations. The numerical results highlight the properties of different gradient compression algorithms and confirm that fast convergence with limited information exchange is possible.

关键词： Convergence Complexity theory Convex functions optimization Quantization (signal) Heuristic algorithms conferences

来源：评论

学校读者我要写书评

暂无评论

Dynamic optimization with CasADi

Dynamic optimization with CasADi

引用

51st IEEE Annual conference on Decision and Control (CDC)

作者： Andersson, Joel Akesson, Johan Diehl, Moritz Katholieke Univ Leuven Elect Engn Dept ESAT Kasteelpk Arenberg 10 B-3001 Heverlee Belgium

ISBN: (纸本)9781467320665

We demonstrate how CasADi, a recently developed, free, open-source, general purpose software tool for nonlinear optimization, can be used for dynamic optimization in a flexible, interactive and numerically efficient way. CasADi is best described as a minimalistic computer algebra system (CAS) implementing automatic differentiation (AD) in eight different flavors. Similar to algebraic modeling languages like AMPL or GAMS, it includes high-level interfaces to state-of-the-art numerical codes for nonlinear programming, quadratic programming and integration of differential-algebraic equations. CasADi is implemented in self-contained C++ code and contains full-featured front-ends to Python and Octave for rapid prototyping. In this paper, we discuss CasADi from the perspective of the developer or advanced user of algorithms for dynamic optimization for the first time, leaving out details on the implementation of the tool. We demonstrate how the tool can be used to model highly complex dynamical systems directly or import existing models formulated in the algebraic modeling language AMPL or the physical modeling language Modelica. Given this symbolic representation of the process models, the resulting optimal control problem can be solved using a variety of methods, including transcription methods (collocation), methods with embedded integrators (multiple shooting) as well as indirect methods.

关键词： C++ language differentiation dynamic programming integration interactive systems public domain software quadratic programming symbol manipulation

来源：评论

学校读者我要写书评

暂无评论

Bio-Inspired algorithms for Prey Model optimization(February 2022) 2

Bio-Inspired Algorithms for Prey Model Optimization(February...

引用

2nd International conference on Innovative Practices in Technology and Management, ICIPTM 2022

作者： Rastogi, Ashish Kumar Taterh, Swapnesh Suresh Kumar, B. Amity University Jaipur India Amity Institute Of Information Technology Amity University Rajasthan India Sanjay Ghodawat University Kolhapur India

ISBN: (纸本)9781665466431

optimization means making the best use of anything or using resources effectively or achieving high quality under the constraints offered. Or at the lowest cost to achieve the optimum result. This research paper contains the algorithms all of which are inspired by the design of BAT algorithms, firefly algorithms and cuckoo algorithms. Also their improved version to optimize the various problems for example real life problems like the traveling salesman problem, manufacturing process optimization, power system stabilizer, engineering optimization etc. BAT algorithm is a type of algorithm which is on the echolocation behavior of bat and this algorithm is Meta heuristic in nature. The algorithms for the optimizations are highly inspired by the nature, behavior of the different animals, birds, wolves, fireflies, fishes etc. The old methods are actually not very efficient as they have some restrictions while these evolutionary algorithms are solving these problems and finding the optimum solution. optimization problems are very challenging to solve either those are of single objective or of the multi objective. Also it is said that NP-hard is not possible to solve this it efficiently by any algorithm in an acceptable time period. As there were real world problems which are actually nonlinear. As these problem leads to the complexity, so to solve these types of problems is not an easy task done by any software. © 2022 IEEE.

关键词： Traveling salesman problem

来源：评论

学校读者我要写书评

暂无评论

Comparing the performance of Moga and MOPSO in optimization of the HEPS performance 7

Comparing the performance of Moga and MOPSO in optimization ...

引用

7th International Particle Accelerator conference, IPAC 2016

作者： Jiao, Y. Xu, G. Key Laboratory of Particle Acceleration Physics and Technology Institute of High Energy Physics Beijing100049 China

ISBN: (纸本)9783954501472

The high Energy Photon Source (HEPS), a kilometrescale diffraction-limited storage ring light source, with a beam energy of 5 to 6 GeV and transverse emittances of a few tens of ***, is to be built in Beijing. A preliminary design with a hybrid 7BA lattice, a natural emittance of 60 *** and a circumference of about 1.3 kilometers, has been made. Based on this design, we optimized the linear and nonlinear performance of the ring with the MOGA and MOPSO algorithms. From comparison of the performance of these two algorithms, it was found that MOPSO promises higher diversity than MOGA, while MOGA can reach better convergence than MOPSO. To reach a true Pareto front, a successive and iterative implementation of the PSO and MOGA, rather than using either of these two algorithms, is suggested. Copyright © 2016 CC-BY-3.0 and by the respective authors.

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：