检索结果-内蒙古大学图书馆

Statistical comparisons of Classifiers by Generalized Stochastic Dominance

JOURNAL OF MACHINE LEARNING RESEARCH 2023年第1期24卷 1-37页

作者： Jansen, Christoph Nalenz, Malte Schollmeyer, Georg Augustin, Thomas Ludwig Maximilians Univ Munchen Dept Stat Ludwigstr 33 D-80539 Munich Germany

Although being a crucial question for the development of machine learning algorithms, there is still no consensus on how to compare classifiers over multiple data sets with respect to several criteria. Every comparison framework is confronted with (at least) three fundamental challenges: the multiplicity of quality criteria, the multiplicity of data sets and the randomness of the selection of data sets. In this paper, we add a fresh view to the vivid debate by adopting recent developments in decision theory. Based on so-called preference systems, our framework ranks classifiers by a generalized concept of stochastic dominance, which powerfully circumvents the cumbersome, and often even self-contradictory, reliance on aggregates. Moreover, we show that generalized stochastic dominance can be operationalized by solving easy-to-handle linear programs and moreover statistically tested employing an multiple quality criteria simultaneously. We illustrate and investigate our framework in a simulation study and with a set of standard benchmark data sets.

关键词： algorithm comparison statistical test generalized stochastic dominance preference system decision theory

来源：评论

学校读者我要写书评

暂无评论

Using the IPOL Journal for Online Reproducible Research in Remote Sensing

引用

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 2020年 13卷 6384-6390页

作者： Colom, Miguel Dagobert, Tristan de Franchis, Carlo von Gioi, Rafael Grompone Hessel, Charles Morel, Jean-Michel Univ Paris Saclay Cole Normale Supr Paris Saclay CNRS F-91190 Gif Sur Yvette France

Reproducible research is needed to ensure that scientific results in the literature are reliable, unbiased, and verifiable by others. The journal Image Processing On Line (IPOL) publishes reproducible articles since 2010. This means publishing an algorithm by a literary description, a pseudocode, its source code, a series of test examples, an online facility allowing to test the code on this data and other data submitted by the user, and finally an experimental archive. In this work, we discuss how to publish and review reproducible research in the specific discipline of remote sensing. We put a special emphasis on the construction and proper documentation of public datasets. We show case studies of remote sensing articles publicly available in IPOL, which demonstrate the feasibility of reproducible research in this area. The methods and their application are explained, along with details on how the datasets were built and made available for evaluation, comparison, and scoring to eventually help establish a reliable state-of-the-art of the discipline. Finally, we give specific recommendations for authors and editors willing to publish reproducible research in remote sensing.

关键词： Remote sensing Reliability Image processing Earth Scholarships Software Licenses algorithm comparison datasets remote sensing reproducible research image processing on line (IPOL)

来源：评论

学校读者我要写书评

暂无评论

Routing algorithm selection for field coverage planning based on field shape and fleet size

引用

COMPUTERS AND ELECTRONICS IN AGRICULTURE 2019年 156卷 523-529页

作者： Seyyedhasani, Hasan Dvorak, Joseph S. Roemmele, Eric Univ Calif Davis Biol & Agr Engn 2030 Bainer Hall Davis CA 95616 USA Univ Kentucky Biosyst & Agr Engn Dept 128 CE Barnhart Bldg Lexington KY 40546 USA Univ Kentucky Appl Stat Lab 323 Plant Sci Bldg Lexington KY 40546 USA

This project considered the field efficiency of routes created using two routing algorithms, a fast Clarke-Wright heuristic and a slower tabu search meta-heuristic, in 100 different field shapes with 1200 different field work scenarios for each algorithm. Different field work scenarios were generated by varying the number of simultaneously working vehicles (1, 2, 3, and 5) and using three path generation methods. After calculating the routes for all the scenarios using cloud computing and a high-performance compute cluster, the routes were evaluated to determine the field efficiency and field completion time. All tabu search results had field efficiencies above 0.65 and field completion times tightly clustered based on the number of simultaneously working vehicles. Many Clarke-Wright results were clustered with the tabu search results, but others had field efficiencies below 0.37 which indicated that the method failed to produce acceptable results. A logistic regression model was developed to identify the factors that caused unacceptable Clarke-Wright results. This enabled creation of a probability equation for predicting when the fast Clarke-Wright equation would produce acceptable results, and conversely, when it would be necessary to rely on the slower tabu search routing. Field shape complexity as measured by the isoperimetric quotient was the primary indicator that Clarke-Wright could fail to produce an acceptable result. The number of simultaneously operating vehicles was statistically insignificant for one, two or three vehicle fleets, but increasing to five vehicles did adversely affect the ability of the Clarke-Wright algorithm to provide acceptable routes.

关键词： Vehicle routing problem Tabu search Clarke-wright algorithm comparison Agricultural machinery

来源：评论

学校读者我要写书评

暂无评论

A survey on evaluation methods for image segmentation

引用

PATTERN RECOGNITION 1996年第8期29卷 1335-1346页

作者： Zhang, YJ SHANTOU UNIV INST AI & PRSHANTOU 515063PEOPLES R CHINA

This paper studies different methods proposed so far for segmentation evaluation. Most methods can be classified into three groups: the analytical, the empirical goodness and the empirical discrepancy groups. Each group has its own characteristics. After a brief description of each method in every group, some comparative discussions about different method groups are first carried out. An experimental comparison for some empirical (goodness and discrepancy) methods commonly used is then performed to provide a rank of their evaluation abilities. In addition, some special methods are also discussed. This study is helpful for an appropriate use of existing evaluation methods and for improving their performance as well as for systematically designing new evalution methods. Copyright (C) 1996 Pattern Recognition Society. Published by Elsevier Science Ltd.

关键词： image analysis image segmentation segmentation evaluation analytical and empirical study performance assessment criteria function algorithm comparison image quality measure method characterization

来源：评论

学校读者我要写书评

暂无评论

Ethylene-vinyl acetate foam as a new lung substitute in radiotherapy

引用

MEDICAL PHYSICS 2018年第4期45卷 1715-1723页

作者： Marques, Enrique Mancha, Pedro J. Virgen del Puerto Hosp Dept Med Phys Plasencia 10600 Caceres Spain Infanta Cristina Hosp Dept Med Phys Badajoz 06006 Spain

PurposeThe purpose of this study was to evaluate ethylene-vinyl acetate (EVA) foam as a new lung substitute in radiotherapy and to study its physical and dosimetric characteristics. MethodsWe calculated the ideal vinyl acetate (VA) content of EVA foam sheets to mimic the physical and dosimetric characteristics of the ICRU lung tissue. We also computed the water-to-medium mass collision stopping power ratios, mass attenuation coefficients, CT numbers, effective atomic numbers and electron densities for: ICRU lung tissue, the RANDO commercial phantom, scaled WATER and EVA foam sheets with varying VA contents in a range between the minimum and maximum values supplied by the manufacturer. For all these substitutes, we simulated percent depth-dose curves with EGSnrc Monte Carlo (MC PDDs) in a water-lung substitute-water slab phantom expressed as dose-to-medium and dose-to-water for 3x3- and 10x10-cm(2) field sizes. PDD for the 10x10-cm(2) field size was also calculated with the MultiGrid Superposition algorithm (MGS PDD) for a relative electron density to water ratio of 0.26. The latter was compared with the MC PDDs in dose-to-water for scaled WATER and EVA foam sheets with the VA content that was most similar to the calculated ideal content that is physically achievable in practice. ResultsWe calculated an ideal VA content of 55%;however, the maximum physically achievable content with current manufacturing techniques is 40%. The physical characteristics of the EVA foam sheets with a VA content of 40% (EVA40) are very close to those of the ICRU lung reference. The physical densities of the EVA40 foam sheets ranged from 0.030 to 0.965g/cm(3), almost covering the entire physical density range of the inflated/deflated lung (0.260-1.050g/cm(3)). Its mass attenuation coefficient at the effective energy of a 6-MV photon beam agrees within 0.8% of the ICRU reference value, and its CT number agrees within 6HU. The effective atomic number for EVA40 varies by less than 0.42 of the

关键词： algorithm comparison dosimetric characteristics ethylene-vinyl acetate foam lung substitute physical characteristics

来源：评论

学校读者我要写书评

暂无评论

Using Well-Understood Single-Objective Functions in Multiobjective Black-Box Optimization Test Suites

引用

EVOLUTIONARY COMPUTATION 2022年第2期30卷 165-193页

作者： Brockhoff, Dimo Auger, Anne Hansen, Nikolaus Tusar, Tea Inst Polytech Paris Ecole Polytech CNRS InriaCMAP Paris France Jozef Stefan Inst Ljubljana Slovenia

Several test function suites are being used for numerical benchmarking of multiobjective optimization algorithms. While they have some desirable properties, such as well-understood Pareto sets and Pareto fronts of various shapes, most of the currently used functions possess characteristics that are arguably underrepresented in real-world problems such as separability, optima located exactly at the boundary constraints, and the existence of variables that solely control the distance between a solution and the Pareto front. Via the alternative construction of combining existing single-objective problems from the literature, we describe the bbob-biobj test suite with 55 bi-objective functions in continuous domain, and its extended version with 92 bi-objective functions (bbob-biobj-ext). Both test suites have been implemented in the COCO platform for black-box optimization benchmarking and various visualizations of the test functions are shown to reveal their properties. Besides providing details on the construction of these problems and presenting their (known) properties, this article also aims at giving the rationale behind our approach in terms of groups of functions with similar properties, objective space normalization, and problem instances. The latter allows us to easily compare the performance of deterministic and stochastic solvers, which is an often overlooked issue in benchmarking.

关键词： Black-box optimization benchmarking multiobjective optimization algorithm comparison benchmark suite generator

来源：评论

学校读者我要写书评

暂无评论

A hybrid swarm-based algorithm for single-objective optimization problems involving high-cost analyses

引用

SWARM INTELLIGENCE 2016年第2期10卷 99-121页

作者： Ampellio, Enrico Vassio, Luca Politecn Torino Dipartimento Ingn Meccan & Aerosp I-10129 Turin TO Italy Politecn Torino Dipartimento Elettron & Telecomunicaz I-10129 Turin TO Italy

In many technical fields, single-objective optimization procedures in continuous domains involve expensive numerical simulations. In this context, an improvement of the Artificial Bee Colony (ABC) algorithm, called the Artificial super-Bee enhanced Colony (AsBeC), is presented. AsBeC is designed to provide fast convergence speed, high solution accuracy and robust performance over a wide range of problems. It implements enhancements of the ABC structure and hybridizations with interpolation strategies. The latter are inspired by the quadratic trust region approach for local investigation and by an efficient global optimizer for separable problems. Each modification and their combined effects are studied with appropriate metrics on a numerical benchmark, which is also used for comparing AsBeC with some effective ABC variants and other derivative-free algorithms. In addition, the presented algorithm is validated on two recent benchmarks adopted for competitions in international conferences. Results show remarkable competitiveness and robustness for AsBeC.

关键词： Modified Artificial Bee Colony Engineering optimization Interpolation strategies algorithm comparison

来源：评论

学校读者我要写书评

暂无评论

Material analysis and big data monitoring of sports training equipment based on machine learning algorithm

引用

NEURAL COMPUTING & APPLICATIONS 2022年第4期34卷 2749-2763页

作者： Zhang, Lei Li, Ning Lanzhou Univ Technol Dept Phys Educ Lanzhou 730050 Gansu Peoples R China Hunan Univ Finance & Econ PE Dept Changsha 410205 Hunan Peoples R China

Different machine learning algorithms predict the application effect of perovskite materials in sports training equipment. The sensitivity to material data is different on different ranges of data sets. Therefore, the algorithm needs to be selected according to specific material data samples. This study compares the prediction performance of neural network prediction algorithm (NN), genetic algorithm, and support vector machine-based machine learning algorithm (SVM) and uses statistical analysis to perform data analysis and draw corresponding curves. Moreover, this study uses a single perovskite material to verify the algorithm performance. In addition, based on the real data, the three machine learning algorithms of this study are applied to the related performance prediction, and the comparative analysis method is used to analyze the prediction performance of the machine learning algorithm. Through data analysis and chart analysis, we can see that machine learning algorithms have a certain effect in the application prediction of perovskite materials in sports training equipment. Among the three machine learning algorithms selected in this study, the performance of the machine learning algorithm based on support vector machine in all aspects is more excellent.

关键词： Machine learning algorithm comparison Perovskite Material properties Prediction Physical training

来源：评论

学校读者我要写书评

暂无评论

Best practices for comparing optimization algorithms

引用

OPTIMIZATION AND ENGINEERING 2017年第4期18卷 815-848页

作者： Beiranvand, Vahid Hare, Warren Lucet, Yves Univ British Columbia Dept Comp Sci Kelowna BC Canada Univ British Columbia Dept Math Kelowna BC Canada

Comparing, or benchmarking, of optimization algorithms is a complicated task that involves many subtle considerations to yield a fair and unbiased evaluation. In this paper, we systematically review the benchmarking process of optimization algorithms, and discuss the challenges of fair comparison. We provide suggestions for each step of the comparison process and highlight the pitfalls to avoid when evaluating the performance of optimization algorithms. We also discuss various methods of reporting the benchmarking results. Finally, some suggestions for future research are presented to improve the current benchmarking process.

关键词： Benchmarking algorithm comparison Guidelines Performance Software Testing Metric Timing Optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Comparing algorithms for automated vessel segmentation in computed tomography scans of the lung: the VESSEL12 study

引用

MEDICAL IMAGE ANALYSIS 2014年第7期18卷 1217-1232页

作者： Rudyanto, Rina D. Kerkstra, Sjoerd van Rilowort, Eva M. Fetita, Catalin Brillet, Pierre-Yves Lefevre, Christophe Xue, Wenzhe Zhu, Xiangjun Liang, Jianming Oksuz, Ilkay Unay, Devrim Kadipasoglu, Kamuran Estepar, Raul San Jose Ross, James C. Washko, George R. Prieto, Juan-Carlos Hernandez Hoyos, Marcela Orkisz, Maciej Meine, Hans Huellebrand, Markus Stoecker, Christina Mir, Fernando Lopez Naranjo, Valery Villanueva, Eliseo Staring, Marius Xiao, Changyan Stoel, Berend C. Fabijanska, Anna Smistad, Erik Elster, Anne C. Lindseth, Frank Foruzan, Amir Hossein Kiros, Ryan Popuri, Karteek Cobzas, Dana Jimenez-Carretero, Daniel Santos, Andres Ledesma-Carbayo, Maria J. Helmberger, Michael Urschler, Martin Pienn, Michael Bosboom, Dennis G. H. Campo, Arantza Prokop, Mathias de Jong, Pim A. Ortiz-de-Solorzano, Carlos Munoz-Barrutia, Arrate van Ginneken, Bram Univ Navarra Ctr Appl Med Res E-31080 Pamplona Spain Radboud Univ Nijmegen Med Ctr Diagnost Image Anal Grp NL-6525 ED Nijmegen Netherlands Inst SudParis Telecom Paris France Arizona State Univ Tempe AZ 85287 USA Bahcesehir Univ Istanbul Turkey Brigham & Womens Hosp Boston MA 02115 USA Univ Lyon CREATIS Lyon France Univ Los Andes Bogota Colombia Fraunhofer MEVIS Bremen Germany Leiden Univ Med Ctr Div Image Proc LKEB NL-2300 RA Leiden Netherlands Hunan Univ Changsha Hunan Peoples R China Lodz Univ Technol Inst Appl Comp Sci Lodz Poland Norwegian Univ Sci & Technol Oslo Norway Shahed Univ Tehran Iran Univ Alberta Edmonton AB T6G 2M7 Canada Univ Politecn Madrid E-28040 Madrid Spain CIBER BBN Zaragoza Spain Graz Univ Technol Inst Comp Vis & Graph A-8010 Graz Austria Ludwig Boltzmann Inst Clin Forens Imaging Graz Austria Ludwig Boltzmann Inst Lung Vasc Res Graz Austria Univ Navarra Univ Navarra Clin Dept Pulm E-31080 Pamplona Spain Univ Med Ctr Dept Radiol Utrecht Netherlands

The VESSEL12 (VESsel SEgmentation in the Lung) challenge objectively compares the performance of different algorithms to identify vessels in thoracic computed tomography (CT) scans. Vessel segmentation is fundamental in computer aided processing of data generated by 3D imaging modalities. As manual vessel segmentation is prohibitively time consuming, any real world application requires some form of automation. Several approaches exist for automated vessel segmentation, but judging their relative merits is difficult due to a lack of standardized evaluation. We present an annotated reference dataset containing 20 CT scans and propose nine categories to perform a comprehensive evaluation of vessel segmentation algorithms from both academia and industry. Twenty algorithms participated in the VESSEL12 challenge, held at International Symposium on Biomedical Imaging (ISBI) 2012. All results have been published at the VESSEL12 website http://***. The challenge remains ongoing and open to new participants. Our three contributions are: (1) an annotated reference dataset available online for evaluation of new algorithms;(2) a quantitative scoring system for objective comparison of algorithms;and (3) performance analysis of the strengths and weaknesses of the various vessel segmentation methods in the presence of various lung diseases. (C) 2014 Elsevier B.V. All rights reserved.

关键词： Thoracic computed tomography Lung vessels algorithm comparison Segmentation Challenge

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：