检索结果-内蒙古大学图书馆

A three-stage optimization method for the classroom envelope in primary and secondary schools in China

JOURNAL OF BUILDING ENGINEERING 2022年 52卷

作者： Xu, Yizhe Yan, Chengchu Pan, Yan Zhao, Ke Li, Mei Zhu, Faxing Jiang, Yanlong Nanjing Tech Univ Coll Urban Construction 200 North Zhongshan Rd Nanjing 210009 Peoples R China Nanjing Univ Aeronaut & Astronaut Key Lab Aircraft Environm Control & Life Support MIIT 29 Yudao St Nanjing 210016 Peoples R China Technol & Higher Educ Inst Hong Kong Fac Sci & Technol Hong Kong Peoples R China Chuzhou Univ Sch Biol & Food Engn Chuzhou 239000 Peoples R China

The design optimization of primary and secondary school classrooms should consider the effects of energy, thermal comfort, and daylighting. This paper proposes a three-stage optimization design method of classroom envelope based on multi-objective optimization to realize the efficient and accurate optimization of primary and secondary school classrooms. The method is divided into three modules: building modeling and analysis, meta-model training and hyperparameter optimization, and multi-objective optimization algorithm comparison. The application effect of electrochromic technology in typical primary and secondary school classrooms is discussed through sensitivity analysis, and an improved algorithm SEGA-Q based on GA is proposed. In addition, the performance of three typical multi-objective optimization algorithms in three objective optimizations (i.e., energy, thermal comfort, and daylighting) is compared. Finally, two typical school classrooms are taken as cases to verify the effectiveness of the proposed optimization method. The results show that the application significance of the electrochromic window indicates skylights > south side window > north side window. The SEGA-Q algorithm has more accuracy, stability, and convergence speed advantages in meta-model training and hyperparametric optimization, but its calculation time increases significantly. Among the three multiobjective optimization algorithms, the NSGA-II algorithm has the most significant advantages in most aspects. In addition, compared with the benchmark schemes, the optimized common classroom can reduce the TES by 692.5 kWh at most and increase the PT by 41.9% and UDINO by 57.4%. The optimized classroom with skylight can reduce TES by 1261.9 kWh at most and increase PT by 26% and UDINO by 78.5%.

关键词： Envelope optimization Hybrid algorithm MFNN School classroom 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

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

Maximum number of generations as a stopping criterion considered harmful

引用

APPLIED SOFT COMPUTING 2022年第0期128卷

作者： Ravber, Miha Liu, Shih-Hsi Mernik, Marjan Crepinsek, Matej Univ Maribor Fac Elect Engn & Comp Sci Koroska cesta 46 Maribor 2000 Slovenia Calif State Univ Fresno Dept Comp Sci Fresno CA 93740 USA

Evolutionary algorithms have been shown to be very effective in solving complex optimization problems. This has driven the research community in the development of novel, even more efficient evolutionary algorithms. The newly proposed algorithms need to be evaluated and compared with existing state-of-the-art algorithms, usually by employing benchmarks. However, comparing evolutionary algorithms is a complicated task, which involves many factors that must be considered to ensure a fair and unbiased comparison. In this paper, we focus on the impact of stopping criteria in the comparison process. Their job is to stop the algorithms in such a way that each algorithm has a fair opportunity to solve the problem. Although they are not given much attention, they play a vital role in the comparison process. In the paper, we compared different stopping criteria with different settings, to show their impact on the comparison results. The results show that stopping criteria play a vital role in the comparison, as they can produce statistically significant differences in the rankings of evolutionary algorithms. The experiments have shown that in one case an algorithm consumed 50 times more evaluations in a single generation, giving it a considerable advantage when max gen was used as the stopping criterion, which puts the validity of most published work in question. (C) 2022 The Author(s). Published by Elsevier B.V.

关键词： Evolutionary algorithms Stopping criteria Benchmarking algorithm termination algorithm comparison

来源：评论

学校读者我要写书评

暂无评论

Machine learning exhibited excellent advantages in the performance simulation and prediction of free water surface constructed wetlands

引用

JOURNAL OF ENVIRONMENTAL MANAGEMENT 2022年 309卷 114694-114694页

作者： Guo, Changqiang Cui, Yuanlai Chinese Acad Sci Nanjing Inst Geog & Limnol Key Lab Watershed Geog Sci Nanjing 210008 Peoples R China Changjiang River Sci Res Inst Changjiang Water Re Key Lab Basin Water Resources & Ecoenvironm Sci H Wuhan 430010 Peoples R China Wuhan Univ State Key Lab Water Resource & Hydropower Engn Sc Wuhan 430072 Peoples R China

Optimizing the design and operation parameters of free water surface constructed wetlands (FWS CWs) in runoff regulation and wastewater treatment is necessary to improve the comprehensive performance. In this study, nine machine learning (ML) algorithms were successfully developed to optimize the parameter combinations for FWS CWs. The scale effect of surface area on wetland performance was determined based on consistently smaller predictions (-6.2% to -28.9%) of the nine well-established ML algorithms. The models most suitable for FWS CW performance simulation and prediction were random forest and extra trees algorithms because of their high R-2 values (0.818 in both) with the training set and low mean absolute relative errors (4.7% and 3.8%, respectively) with the test set. Results from feature analysis of the six tree-based algorithms emphasized the importance of water depth and layout of inlet and outlet, and revealed the negligible effect of the aspect ratio. Feature importance and partial dependence analysis enhanced the interpretability of the tree-based algorithms. The proposed ML algorithms enabled the implementation of an extended scenario at a low cost in real time. Therefore, ML algorithms are suitable for expressing the complex and uncertain effects of the design and operation parameters on the performance of FWS CWs. Acquiring datasets consisting of more extensive, uniform, and unbiased parameter combinations is crucial for developing more robust and practical ML algorithms for the optimal design of FWS CWs.

关键词： Wetland performance algorithm comparison Scenario prediction Feature importance Interpretability

来源：评论

学校读者我要写书评

暂无评论

ADMET Predictability at Boehringer Ingelheim: State-of-the-Art, and Do Bigger Datasets or algorithms Make a Difference?

引用

MOLECULAR INFORMATICS 2022年第2期41卷 2100113-2100113页

作者： Aleksic, Stevan Seeliger, Daniel Brown, J. B. Boehringer Ingelheim Pharma GmbH & Co KG Med Chem D-88397 Biberach Germany

Computational methods assisting drug discovery and development are routine in the pharmaceutical industry. Digital recording of ADMET assays has provided a rich source of data for development of predictive models. Despite the accumulation of data and the public availability of advanced modeling algorithms, the utility of prediction in ADMET research is not clear. Here, we present a critical evaluation of the relationships between data volume, modeling algorithm, chemical representation and grouping, and temporal aspect (time sequence of assays) using an in-house ADMET database. We find no large difference in prediction algorithms nor any systemic and substantial gain from increasingly large datasets. Temporal-based data enlargement led to performance improvement in only in a limited number of assays, and with fractional improvement at best. Assays that are well-, intermediately-, or poorly-suited for ADMET predictions and reasons for such behavior are systematically identified, generating realistic expectations for areas in which computational models can be used to guide decision making in molecular design and development.

关键词： ADMET modelling machine learning algorithm comparison chemical representation congeneric series

来源：评论

学校读者我要写书评

暂无评论

An Improved Cross-Validated Adversarial Validation Method 1

引用

16th International Conference on Knowledge Science, Engineering and Management (KSEM)

作者： Zhang, Wen Liu, Zhengjiang Xue, Yan Wang, Ruibo Cao, Xuefei Li, Jihong Shanxi Univ Sch Automat & Software Engn Taiyuan 030006 Peoples R China Shanxi Univ Sch Comp & Informat Technol Taiyuan 030006 Peoples R China Shanxi Univ Sch Modern Educ Technol Taiyuan 030006 Peoples R China

ISBN: (数字)9783031402838

ISBN: (纸本)9783031402821;9783031402838

As a widely-used strategy among Kaggle competitors, adversarial validation provides a novel selection framework of a reasonable training and validation sets. An adversarial validation heavily depends on an accurate identification of the difference between the distributions of the training and test sets released in a Kaggle competition. However, the typical adversarial validation merely uses a K-fold cross-validated point estimator to measure the difference regardless of the variation of the estimator. Therefore, the typical adversarial validation tends to produce unpromising false positive conclusions. In this study, we reconsider the adversarial validation from a perspective of algorithm comparison. Specifically, we formulate the adversarial validation into a comparison task of a well-trained classifier with a random-guessing classifier on an adversarial data set. Then, we investigate the state-of-the-art algorithm comparison methods to improve the adversarial validation method for reducing false positive conclusions. We conducted sufficient simulated and real-world experiments, and we showed the recently-proposed 5 x 2 BCV McNemar's test can significantly improve the performance of the adversarial validation method.

关键词： Adversarial Validation Cross Validation algorithm comparison Significance Testing Distribution Shift

来源：评论

学校读者我要写书评

暂无评论

CONSTRAINED BAYESIAN OPTIMIZATION METHODS USING REGRESSION AND CLASSIFICATION GAUSSIAN PROCESSES AS CONSTRAINTS

CONSTRAINED BAYESIAN OPTIMIZATION METHODS USING REGRESSION A...

引用

ASME International Design Engineering Technical Conferences / Computers and Information in Engineering Conference (IDETC-CIE) / 49th Design Automation Conference (DAC)

作者： Jetton, Cole Li, Chengda Hoyle, Christopher Oregon State Univ Design Engn Lab Corvallis OR 97331 USA

ISBN: (纸本)9780791887318

In Bayesian optimization, which is used for optimizing computationally expensive black-box problems, constraints need to be considered to arrive at solutions that are both optimal and feasible. Techniques to deal with black-box constraints have been studied extensively, often exploiting the constraint model to sample points that have a high probability of being feasible. These constraints are usually modeled using regression models if the output is a known value on a continuous real scale. However, if the constraint or set of constraints is not modeled explicitly, but rather it is only known whether a design vector is either feasible or infeasible, then we treat the constraints as a categorical constraint. Because of this, these constraints should be modeled using classification models rather than regression methods, which have also been studied. Because of the variety of approaches to handling constraints, there is a need to compare methods for handling both classification constraints as well as continuous constraints modeled with individual regression models. This paper explores and compares four main methods, with two additional ones specifically for classification constraints;these methods handle black-box constraints in Bayesian optimization by modeling the constraints with both regression and classification Gaussian processes. We found that the problem type and number of constraints influence the effectiveness of different approaches with statistical differences in convergence. Regression models can be advantageous in terms of model fit time;however, this is also a function of the total number of constraints. Generally, regression constraints outperformed classification surrogates in terms of minimizing computational time, but the latter was still effective. Overall, this study provides valuable insights into the performance and implementation of different constrained Bayesian optimization techniques. This can help inform engineers on which method is most suit

关键词： algorithm comparison Bayesian Optimization Constraint Handling Gaussian Process Classification Gaussian Process Regression Optimization.

来源：评论

学校读者我要写书评

暂无评论

Depth Functions for Partial Orders with a Descriptive Analysis of Machine Learning algorithms 13

Depth Functions for Partial Orders with a Descriptive Analys...

引用

13th International Symposium on Imprecise Probabilities - Theories and Applications (ISIPTA)

作者： Blocher, Hannah Schollmeyer, Georg Jansen, Christoph Nalenz, Malte Ludwig Maximilians Univ Munchen Dept Stat Munich Germany

We propose a framework for descriptively analyzing sets of partial orders based on the concept of depth functions. Despite intensive studies of depth functions in linear and metric spaces, there is very little discussion on depth functions for non-standard data types such as partial orders. We introduce an adaptation of the well-known simplicial depth to the set of all partial orders, the union-free generic (ufg) depth. Moreover, we utilize our ufg depth for a comparison of machine learning algorithms based on multidimensional performance measures. Concretely, we analyze the distribution of different classifier performances over a sample of standard benchmark data sets. Our results promisingly demonstrate that our approach differs substantially from existing benchmarking approaches and, therefore, adds a new perspective to the vivid debate on the comparison of classifiers.

关键词： partial orders data depth benchmarking algorithm comparison outlier detection non-standard data

来源：评论

学校读者我要写书评

暂无评论

Semi-supervised Learning in Distributed Split Learning Architecture and IoT Applications 15

Semi-supervised Learning in Distributed Split Learning Archi...

引用

IEEE 15th International Symposium on Autonomous Decentralized Systems (ISADS)

作者： Barhoush, Mahdi Ayad, Ahmad Schmeink, Anke Rhein Westfal TH Aachen Chair Informat Theory & Data Analyt Aachen Germany

ISBN: (纸本)9781665464512

In the era of big data, new learning techniques are emerging to solve the difficulties of data collection, storage, scalability, and privacy. To overcome these challenges, we propose a distributed learning system that merges the hybrid edge-cloud split-learning architecture with the semi-supervised learning scheme. The proposed system based on three semi-supervised learning algorithms (FixMatch, Virtual Adversarial Training, and MeanTeacher) is compared to the supervised learning scheme and trained on different datasets and data distributions (IID and non-IID) and with a variable number of clients. The new system could efficiently utilize the local unlabeled samples on the client side and gave a performance encouragement that exceeds 30% in most cases even with small percentage of labelled data. Additionally, certain Split-SSL algorithms showed performance that was on par with or occasionally even better than more resource-intensive algorithms, although requiring less processing power and convergence time.

关键词： algorithm comparison Big Data Distributed Learning Semi-supervised learning Split Learning

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：