检索结果-内蒙古大学图书馆

Automated Benchmark-Driven Design and Explanation of Hyperparameter Optimizers

IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION 2022年第6期26卷 1336-1350页

作者： Moosbauer, Julia Binder, Martin Schneider, Lennart Pfisterer, Florian Becker, Marc Lang, Michel Kotthoff, Lars Bischl, Bernd Ludwig Maximilians Univ Munchen Dept Stat D-80539 Munich Germany Res Ctr Trustworthy Data Sci & Secur D-44227 Dortmund Germany Univ Wyoming Dept Comp Sci Laramie WY 82071 USA

Automated hyperparameter optimization (HPO) has gained great popularity and is an important component of most automated machine learning frameworks. However, the process of designing HPO algorithms is still an unsystematic and manual process: new algorithms are often built on top of prior work, where limitations are identified and improvements are proposed. Even though this approach is guided by expert knowledge, it is still somewhat arbitrary. The process rarely allows for gaining a holistic understanding of which algorithmic components drive performance and carries the risk of overlooking good algorithmic design choices. We present a principled approach to automated benchmark-driven algorithm design applied to multifidelity HPO (MF-HPO). First, we formalize a rich space of MF-HPO candidates that includes, but is not limited to, common existing HPO algorithms and then present a configurable framework covering this space. To find the best candidate automatically and systematically, we follow a programming-by-optimization approach and search over the space of algorithm candidates via Bayesian optimization. We challenge whether the found design choices are necessary or could be replaced by more naive and simpler ones by performing an ablation analysis. We observe that using a relatively simple configuration (in some ways, simpler than established methods) performs very well as long as some critical configuration parameters are set to the right value.

关键词： algorithm analysis algorithm design automated machine learning (AutoML) hyperparameter optimization (HPO) multifidelity

来源：评论

学校读者我要写书评

暂无评论

Image segmentation algorithm based on partial differential equation

引用

JOURNAL OF INTELLIGENT & FUZZY SYSTEMS 2021年第4期40卷 5945-5952页

作者： Cui, Di Ningbo Univ Technol Sch Elect & Informat Engn Ningbo Peoples R China

Image segmentation is very important for various fields. With the development of computer technology, computer technology has become more and more effective for image segmentation, and it is studied on the basis of partial differential equations. The curve representation method in plane differential geometry is expounded, with the SegNet-v2 segmentation model analyzed and tested in medical image segmentation. The test results show that the partial differential equation image segmentation algorithm can achieve more accurate segmentation, especially in medical image segmentation, which can achieve good results, and it is worth in practice to further promote.

关键词： Partial differential equation image segmentation algorithm analysis

来源：评论

学校读者我要写书评

暂无评论

A comparison of feature description algorithms

引用

OPTIK 2015年第2期126卷 274-278页

作者： Hu, Jinlong Peng, Xianrong Fu, Chengyu Chinese Acad Sci Inst Opt & Elect Chengdu 610209 Peoples R China Grad Univ Chinese Acad Sci Beijing 100049 Peoples R China

This paper reviews and evaluates several state-of-the-art feature description algorithms. The components of each feature description method are analyzed and their applications in dealing with specific challenges are identified. In the paper, we compare state-of-the-art feature description methods including the SIFT, DAISY, HRI-CSLTP, LIOP, MROGH and MRRID with specific measurement regulation. The quantitative comparative results demonstrate these algorithms' applications in different scenes, which provide a certain guide for designing novel feature description algorithms. (C) 2014 Published by Elsevier GmbH.

关键词： Feature description Performance evaluation algorithm analysis

来源：评论

学校读者我要写书评

暂无评论

algorithmIC PHASE-DIAGRAMS

引用

IEEE TRANSACTIONS ON COMPUTERS 1987年第2期36卷 231-233页

作者： HOCKNEY, R NASA LANGLEY RES CTR INST COMP APPLICAT SCI & ENGNHAMPTONVA 23666

algorithmic phase diagrams are a neat and compact representation of the results of comparing the execution time of several algorithms for the solution of the same problem. As an example we show the recent results of Gannon and Van Rosendale on the solution of multiple tridiagonal systems of equations in the form of such diagrams. The act of preparing these diagrams has revealed an unexpectedly complex relationship between the best algorithm and the number and size of the tridiagonal systems, which was not evident from the algebraic formulae in the original paper. Even so, for a particular computer, one diagram suffices to predict the best algorithm for all problems that are likely to be encountered-the prediction being read directly from the diagram without complex calculation.

关键词： algorithm analysis algorithmic phase diagrams characterization of algorithms n parallel algorithms parallel computation

来源：评论

学校读者我要写书评

暂无评论

A heuristic for mining association rules in polynomial time

引用

MATHEMATICAL AND COMPUTER MODELLING 2003年第1-2期37卷 219-233页

作者： Yilmaz, E Triantaphyllou, E Chen, J Liao, TW Louisiana State Univ Dept Ind & Mfg Syst Engn Baton Rouge LA 70803 USA Gen Elect Card Serv Inc Stamford CT 06927 USA

Mining association rules from databases has attracted great interest because of its potentially very practical applications, Given a database, the problem of interest is how to mine association rules (which could describe patterns of consumers' behaviors) in an efficient and effective way, The databases involved in today's business environments can be very large. Thus, fast and effective algorithms are needed to mine association rules out of large databases, Previous approaches may cause an exponential computing resource consumption. A combinatorial explosion occurs because existing approaches exhaustively mine all the rules. The proposed algorithm takes a previously developed approach, called the Randomized algorithm 1 (or RA1), and adapts it to mine association rules out of a database in an efficient way. The original RA1 approach was primarily developed for inferring logical clauses (i.e., a Boolean function) from examples. Numerous computational results suggest that the new approach is very promising. (C) 2003 Elsevier Science Ltd. All rights reserved.

关键词： data mining association rules algorithm analysis one clause at a time (OCAT) approach randomized algorithms heuristics boolean functions

来源：评论

学校读者我要写书评

暂无评论

APPROXIMATION algorithmS FOR SCHEDULING ON UNIFORM PROCESSORS

引用

INFOR 1993年第1期31卷 16-23页

作者： FRACCHIA, FD SAXTON, LV

The problem of scheduling independent tasks on uniform processors in order to minimize the maximum completion time of the schedule is strongly NP-complete. Therefore, the existence of a polynomial or even a pseudo-polynomial time algorithm is very unlikely unless P = NP. This paper deals with approximation schemes which attempt to find a solution in polynomial time that is close to the optimal solution. Specifically, two variants of LPT scheduling are shown: one which is polynomial time and the other a probabilistically good algorithm.

关键词： algorithm analysis APPROXIMATION algorithmS SCHEDULING UNIFORM PROCESSES

来源：评论

学校读者我要写书评

暂无评论

A Cost-Aware Logical Framework

引用

PROCEEDINGS OF THE ACM ON PROGRAMMING LANGUAGES-PACMPL 2022年第POPL期6卷 1–31页

作者： Niu, Yue Sterling, Jonathan Grodin, Harrison Harper, Robert Carnegie Mellon Univ 5000 Forbes Ave Pittsburgh PA 15213 USA Aarhus Univ Aabogade 34 DK-8000 Aarhus C Denmark

We present calf, a cost-aware logical framework for studying quantitative aspects of functional programs. Taking inspiration from recent work that reconstructs traditional aspects of programming languages in terms of a modal account of phase distinctions, we argue that the cost structure of programs motivates a phase distinction between intension and extension. Armed with this technology, we contribute a synthetic account of cost structure as a computational effect in which cost-aware programs enjoy an internal noninterference property: input/output behavior cannot depend on cost. As a full-spectrum dependent type theory, calf presents a unified language for programming and specification of both cost and behavior that can be integrated smoothly with existing mathematical libraries available in type theoretic proof assistants. We evaluate calf as a general framework for cost analysis by implementing two fundamental techniques for algorithm analysis: the method of recurrence relations and physicist's method for amortized analysis. We deploy these techniques on a variety of case studies: we prove a tight, closed bound for Euclid's algorithm, verify the amortized complexity of batched queues, and derive tight, closed bounds for the sequential and parallel complexity of merge sort, all fully mechanized in the Agda proof assistant. Lastly we substantiate the soundness of quantitative reasoning in calf by means of a model construction.

关键词： algorithm analysis cost models phase distinction noninterference intensional property behavioral verification equational reasoning modal type theory mechanized proof proof assistants recurrence relations amortized analysis parallel algorithms

来源：评论

学校读者我要写书评

暂无评论

analysis of Power-Aware Buffering Schemes in Wireless Sensor Networks

引用

ACM TRANSACTIONS ON SENSOR NETWORKS 2010年第3期7卷 1–33页

作者： Ling, Yibei Chen, Chung-Min Chen, Shigang Telcordia Technol Appl Res Piscataway NJ 08854 USA Univ Florida Dept Comp & Informat Sci & Engn Gainesville FL 32611 USA

We study the power-aware buffering problem in battery-powered sensor networks, focusing on the fixed-size and fixed-interval buffering schemes. The main motivation is to address the yet poorly understood size variation-induced effect on power-aware buffering schemes. Our theoretical analysis elucidates the fundamental differences between the fixed-size and fixed-interval buffering schemes in the presence of data-size variation. It shows that data-size variation has detrimental effects on the power expenditure of the fixed-size buffering in general, and reveals that the size variation induced effects can be either mitigated by a positive skewness or promoted by a negative skewness in size distribution. By contrast, the fixed-interval buffering scheme has an obvious advantage of being eminently immune to the data-size variation. Hence the fixed-interval buffering scheme is a risk-averse strategy for its robustness in a variety of operational environments. In addition, based on the fixed-interval buffering scheme, we establish the power consumption relationship between child nodes and parent node in a static data-collection tree, and give an in-depth analysis of the impact of child bandwidth distribution on the parent's power consumption. This study is of practical significance: it sheds new light on the relationship among power consumption of buffering schemes, power parameters of radio module and memory bank, data arrival rate, and data-size variation, thereby providing well-informed guidance in determining an optimal buffer size (interval) to maximize the operational lifespan of sensor networks.

关键词： algorithms Power-aware buffering schemes algorithm analysis

来源：评论

学校读者我要写书评

暂无评论

Average-Case analysis of Some Plurality algorithms

引用

ACM TRANSACTIONS ON algorithmS 2009年第2期5卷 17-17页

作者： Alonso, Laurent Reingold, Edward M. Univ Nancy 1 INRIA Lorraine & LORIA F-54506 Vandoeuvre Les Nancy France IIT Dept Comp Sci Chicago IL 60616 USA

Given a set of n elements, each of which is colored one of c colors, we must determine an element of the plurality (most frequently occurring) color by pairwise equal/unequal color comparisons of elements. We focus on the expected number of color comparisons when the c(n) colorings are equally probable. We analyze an obvious algorithm, showing that its expected performance is c(2) 1c 2/2cn-O(c(2)), with variance Theta(c(2)n). We present and analyze an algorithm for the case c = 3 colors whose average complexity on the 3(n) equally probable inputs is 7083/5425n+O(root n) = 1.3056....n+O(root n), substantially better than the expected complexity 5/3n + O(1) = 1.6666...n + O(1) of the obvious algorithm. We describe a similar algorithm for c = 4 colors whose average complexity on the 4(n) equally probable inputs is 761311/402850n + O(log n) = 1.8898...n+O(log n), substantially better than the expected complexity 9/4n + O(1) = 2.25n + O(1) of the obvious algorithm.

关键词： algorithm analysis majority problem plurality problem

来源：评论

学校读者我要写书评

暂无评论

A note on asynchronous multi-exponentiation algorithm using binary representation

引用

INFORMATION PROCESSING LETTERS 2012年第22期112卷 876-879页

作者： Sun, Da-Zhi Huai, Jin-Peng Li, Jian-Xin Tianjin Univ Sch Comp Sci & Technol Tianjin 300072 Peoples R China Beihang Univ Sch Comp Sci Beijing 100083 Peoples R China

A basic but expensive operation in implementations of several famous public-key crypto-systems is the computation of the multi-exponentiation in a certain finite multiplication group. In 2007, Yang et al. presented an interesting asynchronous multi-exponentiation algorithm called the SUt method, which uses the binary representations for the exponents. In this note, we analyze the computational efficiency of the SUt method by modeling the scanning process as a Markov chain. It shows that their computational efficiency result is incorrect. Moreover, we make a performance comparison among the published techniques and show that the performance of a modified sliding window method is better than that of the SUt method, when there are 4 or more additional registers. We hope that our research will be convenient to the development of the cryptographic devices. (C) 2012 Elsevier By. All rights reserved.

关键词： Cryptography Multi-exponentiation algorithm analysis Markov chain

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：