检索结果-内蒙古大学图书馆

Towards effective metamorphic testing by algorithm stability for linear classification programs

JOURNAL OF SYSTEMS AND SOFTWARE 2021年 180卷 111012-111012页

作者： Yang, Yingzhuo Li, Zenan Wang, Huiyan Xu, Chang Ma, Xiaoxing Nanjing Univ State Key Lab Novel Software Technol Nanjing Peoples R China Nanjing Univ Dept Comp Sci & Technol Nanjing Peoples R China

The quality assurance for machine learning systems is becoming increasingly critical nowadays. While many efforts have been paid on trained models from such systems, we focus on the quality of these systems themselves, as the latter essentially decides the quality of numerous models thus trained. In this article, we focus particularly on detecting bugs in implementing one class of model-training systems, namely, linear classification algorithms, which are known to be challenging due to the lack of test oracle. Existing work has attempted to use metamorphic testing to alleviate the oracle problem, but fallen short on overlooking the statistical nature of such learning algorithms, leading to premature metamorphic relations (MRs) suffering efficacy and necessity issues. To address this problem, we first derive MRs from a fundamental property of linear classification algorithms, i.e., algorithm stability, with the soundness guarantee. We then formulate such MRs in a way that is rare in usage but could be more effective according to our field study and analysis, i.e., Past-execution Dependent MR (PD-MR), as contrast to the traditional way, i.e., Past-execution Independent MR (PI-MR), which has been extensively studied. We experimentally evaluated our new MRs upon nine well-known linear classification algorithms. The results reported that the new MRs detected 37.6-329.2% more bugs than existing benchmark MRs. (C) 2021 Elsevier Inc. All rights reserved.

关键词： Metamorphic testing Metamorphic relation Machine learning program Linear classifier algorithm stability

来源：评论

学校读者我要写书评

暂无评论

UNMANNED AIR VEHICLE 3-D WING AERODYNAMICAL DESIGN AND algorithm stability WITH RESPECT TO INITIAL SHAPE

引用

THERMAL SCIENCE 2019年第Sup2期23卷 S599-S605页

作者： Peigin, Sergey Pushchin, Nikita Timchenko, Sergey OPTIMENGA 777 Ltd Moscow Russia Tomsk State Univ Tomsk Russia

A new technology of the optimal design of aerodynamic configurations based on a new generation software product is used for aerodynamic design of a 3-D wing of the middle class unmanned aerial vehicle. The optimal shape of the wing, which is characterized by minimum total drag at a fixed lift coefficient and corresponding to the specified geometric and aerodynamic constraints, is determined on the basis of the global search method and numerical solutions of the complete Navier-Stokes equations. It is shown that the proposed approach provides reduction in a wing drag in the cruise flight zone and significantly reduces the material and time costs for aerodynamic aircraft design. Optimal wing has a significantly lower drag at the main design point, and it can be used during cruising and in its vicinity. Optimization allows improving of the glider wing quality. Optimal wing is distinguished by better aerodynamic characteristics in the wide vicinity of design point in terms of the Mach numbers and lift coefficient. Such wing is resistant to the small changes in the flight conditions and it meets all given geometric and aerodynamic constraints.

关键词： optimal design full Navier-Stokes equations non-linear constraints drag coefficient pitch moment algorithm stability

来源：评论

学校读者我要写书评

暂无评论

Estimate of stability of the Dynamic Clustering algorithm for Temperature Signals

引用

OPTOELECTRONICS INSTRUMENTATION AND DATA PROCESSING 2020年第6期56卷 580-585页

作者： Volkov, Yu V. Russian Acad Sci Inst Monitoring Climat & Ecol Syst Siberian Branch Tomsk 634055 Russia

A dynamic clustering algorithm for temperature signals is considered. It is used for solving the problem of distinguishing climatic regions characterized by a certain homogeneous climate type within the boundaries of the region on the Earth's surface. A numerical experiment implemented with the aim of estimating its stability is described. The reference signal model used in the numerical experiment is presented as a sum of harmonic components. Additive components of the noise are formed in the frequency region. In the numerical experiment, the noise magnitude and sample size are measured. Estimates of root-mean-square deviations and stability of the dynamic clustering algorithm are presented.

关键词： signal analysis mathematical model surface temperature climate classification clustering algorithm stability

来源：评论

学校读者我要写书评

暂无评论

Eliminating Variable Order Instability in Greedy Score-Based Structure Learning 12

Eliminating Variable Order Instability in Greedy Score-Based...

引用

12th International Conference on Probabilistic Graphical Models (PGM)

作者： Kitson, Neville K. Constantinou, Anthony C. Queen Mary Univ London Machine Intelligence & Decis Syst MInDS Grp Mile End Rd London E1 4NS England

Many Bayesian Network structure learning algorithms are unstable in that the learnt graph is sensitive to arbitrary artefacts of the dataset, such as the ordering of columns (i.e., variable order). PC-Stable, developed by Colombo and Maathuis (2014), attempts to address this issue for the widely-used PC algorithm, prompting researchers to use the 'stable' version instead. However, this problem seems to have been overlooked for score-based algorithms. In this study, we show that some widely-used score-based algorithms suffer from the same issue and that PC-Stable, although less sensitive than most of the score-based algorithms tested, is not completely stable. We also present a solution to score-based greedy hill-climbing that completely eliminates this instability, and provide two implementations: the HC-Stable and Tabu-Stable algorithms, the latter of which learns more accurate graphs than all the well-known algorithms we compared it to.

关键词： Bayesian Networks algorithm stability hill-climbing Tabu algorithm

来源：评论

学校读者我要写书评

暂无评论

Reinforcement Learning using Reward Expectations in Scenarios with Aleatoric Uncertainties

Reinforcement Learning using Reward Expectations in Scenario...

引用

IEEE Conference on Games (CoG)

作者： Wang, Yubin Sun, Yifeng Wu, Jiang Hu, Hao Wu, Zhiqiang Huang, Weigui PLA SSF Informat Engn Univ Zhengzhou Peoples R China

ISBN: (数字)9781665459891

ISBN: (纸本)9781665459891

In scenarios with aleatoric uncertainties, the reward got by an agent when executing the same action in the same state is random, which can reduce the stability and convergence speed of the reinforcement algorithms. However, in most scenarios, reward functions have regularity, and their expectations are determined, which can be got through models or sample statistics. This paper discusses the distribution relationship between reward functions and value functions in scenarios with aleatoric uncertainties and proves the feasibility of using reward expectations for reinforcement learning. Finally, experiments show that algorithms have better stability and convergence speed when using reward expectations than random rewards.

关键词： reinforcement learning aleatoric uncertainty reward expectations algorithm stability

来源：评论

学校读者我要写书评

暂无评论

The generalization error of graph convolutional networks may enlarge with more layers

引用

NEUROCOMPUTING 2021年 424卷 97-106页

作者： Zhou, Xianchen Wang, Hongxia Natl Univ Def Technol Coll Sci Changsha 410073 Hunan Peoples R China

Graph Neural Networks(GNNs) are powerful methods to analyze the non-Euclidean data. As a dominant type of GNN, Graph Convolutional Networks(GCNs) have wide applications. However, the analysis of the generalization error for GCNs with multilayer is limited. Based on the review of single-layer GCNs, this paper analyzes the generalization error of two-layers GCNs and extends the conclusion to the general GCNs models. Firstly, this paper examines two-layers GCNs and obtains the stability of the GCNs algo-rithm. And then, based on this algorithmic stability, the generalization stability of multilayer GCNs is obtained. This paper shows that the algorithmic stability of GCNs depends upon the graph filters and its product with node features as well as the training procedure. Furthermore, the generalization error gap of GCNs tends to be enlarged with more layers, which can interpret why GCNs with deeper layers have relatively poorer performance in test datasets. (c) 2020 Elsevier B.V. All rights reserved.

关键词： Generalization error algorithm stability Graph convolutional network

来源：评论

学校读者我要写书评

暂无评论

Three-way trade-off in multi-objective learning: optimization, generalization and conflict-avoidance

The Journal of Machine Learning Research

引用

The Journal of Machine Learning Research 2024年第1期25卷 9153-9205页

作者： Lisha Chen Heshan Fernando Yiming Ying Tianyi Chen Department of Electrical Computer & Systems Engineering Rensselaer Polytechnic Institute School of Mathematics and Statistics University of Sydney NSW Australia

Multi-objective learning (MOL) often arises in machine learning problems when there are multiple data modalities or tasks. One critical challenge in MOL is the potential conflict among different objectives during the optimization process. Recent works have developed various dynamic weighting algorithms for MOL, where the central idea is to find an update direction that avoids conflicts among objectives. Albeit its appealing intuition, empirical studies show that dynamic weighting methods may not outperform static ones. To understand this theory-practice gap, we focus on a stochastic variant of MGDA, the Multi-objective gradient with Double sampling (MoDo), and study the generalization performance and its interplay with optimization through the lens of algorithmic stability in the framework of statistical learning theory. We find that the key rationale behind MGDA--updating along conflict-avoidant direction--may hinder dynamic weighting algorithms from achieving the optimal O(1/√n) population risk, where n is the number of training samples. We further demonstrate the impact of dynamic weights on the three-way trade-off among optimization, generalization, and conflict avoidance unique in MOL. We showcase the generality of our theoretical framework by analyzing other algorithms under the framework. Experiments on various multi-task learning benchmarks are performed to demonstrate the practical applicability. Code is available at https://***/heshandevaka/Trade-Off-MOL.

关键词： multi-objective optimization statistical learning theory algorithm stability Pareto stationarity gradient conflict

来源：评论

学校读者我要写书评

暂无评论

Simulation of a direct torque control system in the presence of winding asymmetry in induction motor

引用

ENGINEERING RESEARCH EXPRESS 2023年第2期5卷 025070-025070页

作者： Goolak, Sergey Liubarskyi, Borys Riabov, Ievgen Chepurna, Nataliia Pohosov, Oleksandr Railways State Univ Infrastruct & Technol Dept Electromech & Rolling Stock Kyrylivska Str 9 UA-04071 Kiev Ukraine Natl Tech Univ Dept Elect Transport & Locomot Engn Kharkiv Polytech Inst Kyrpychova Str 2 UA-61002 Kharkiv Ukraine Kyiv Natl Univ Construct & Architecture Dept Heat Engn Povitroflotskiy Ave 31 UA-03037 Kiev Ukraine

The operation of a direct torque control (DTC) system for induction motors with the possible presence of unbalanced modes in the motor windings is studied. It is shown that in the presence of asymmetric modes in the motor windings, the classical DTC algorithm becomes unstable. It is proposed to improve the DTC by taking into account the unbalance of the active resistances of the motor stator windings when determining the DTC parameters. This will make it possible to study electrodynamic processes in DTC with the possible presence of asymmetric modes in the windings of induction motor.

关键词： electric motor stock induction motor traction drive direct torque control asymmetric modes torque pulsations algorithm stability

来源：评论

学校读者我要写书评

暂无评论

Machine learning for urban land use/ cover mapping: Comparison of artificial neural network, random forest and support vector machine, a case study of Dilla town

引用

Heliyon 2024年第20期10卷 e39146页

作者： Kasahun, Melion Legesse, Abiyot Department of Geography and Environmental Studies College of Social Science and Humanities Dilla University P.O. Box: 419 Dilla Ethiopia Department of Geography and Environmental Studies College of Social Science and Humanities Borana University P.O. Box: 85 Yabello Ethiopia

The ability to accurately classify land use/cover (LULC) is critical for environmental monitoring and land use planning. This study compares three machine learning algorithms: Artificial Neural Network (ANN), Support Vector Machine (SVM), and Random Forest (RF) for LULC classification using Google Earth images from the years 2006, 2014, and 2022. The objective of this study is to evaluate and identify the best classifier for LULC classification and change detection. Four LULC categories (Built-up, Open area, Farmland, and Agroforestry) were identified. The evaluation criteria included overall accuracy, kappa coefficient, producer's accuracy, user's accuracy, computing time, algorithm stability, and visual quality. The results showed that the RF algorithm outperformed both SVM and ANN algorithms with an average overall accuracy of 0.97, kappa coefficient of 0.98, producer's accuracy of 0.99, and user's accuracy of 0.97, surpassing the accuracies achieved by SVM (0.96, 0.97, 0.98, and 0.97) and ANN (0.89, 0.81, 0.94, and 0.88), with corresponding computing times of 6.33, 15, and 30 s. All classifiers performed stably with different training sizes. Visual quality assessment revealed that RF had the highest precision. Consequently, the built-up change detection result shows, the net change in built-up area between 2006 and 2022 was increased by 0.74 Km² for ANN, 1.74 Km2 for SVM, and 1.66 Km² for RF. The comparison reveals that the RF algorithm showcasing high precision in detecting change, consistent with the data (increased by 1.65 Km²) obtained from Dilla town land administration office. To validate the results, the study considered field surveys, reference images, local experts, and previous studies. Based on the findings, the study concludes that using RF classifier with an object-based approach is an effective way to map LULC and detect changes in the study area over time. Future researchers are recommended to utilize this effectiv

关键词： Accuracy assessment algorithm stability ANN Computing time LULC RF SVM Visual quality

来源：评论

学校读者我要写书评

暂无评论

Coarse-refinement dilemma: On generalization bounds for data clustering

引用

EXPERT SYSTEMS WITH APPLICATIONS 2021年 184卷 115399-115399页

作者： Vaz, Yule de Mello, Rodrigo Fernandes Grossi Ferreira, Carlos Henrique Univ Sao Paulo Inst Math & Comp Sci Trabalhador Saocarlense Ave 400 BR-13560970 Sao Carlos SP Brazil

The data clustering problem is of central importance for the area of machine learning, given its usefulness to represent data structural similarities from input spaces. Although, data clustering counts on scarse literature of a theoretical framework with generalization guarantees. In this context, this manuscript introduces a new concept, based on multidimensional persistent homology, to analyze the conditions on which a clustering model is capable of generalizing data. As a first step, we propose a more general definition of DC problem by relying on topological spaces, instead of metric ones as typically approached in the literature. From that, we show that the data clustering problem presents an analogous dilemma to the bias-variance one, which is here referred to as the coarse-refinement dilemma, from which we conclude that: (i) highly-refined partitions and the clustering instability (overfitting);and (ii) over-coarse partitions and the lack of representativeness (underfitting). The coarse-refinement dilemma suggests the need of a relaxation of Kleinberg's richness axiom, as such axiom allows the production of unstable or unrepresentative partitions. Experimental exploration considering different clustering refinements can, then, depict such partitions.

关键词： Data clustering Topology Persistent homology Multidimensional persistence algorithm stability

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：