检索结果-内蒙古大学图书馆

Patient-Specific hyperparameter optimization of a Deep Learning-Based Tumor Autocontouring Algorithm on 2D Liver, Prostate, and Lung Cine MR Images: A Pilot Study

引用

ALGORITHMS 2025年第4期18卷 233-233页

作者： Han, Gawon Wachowicz, Keith Usmani, Nawaid Yee, Don Wong, Jordan Elangovan, Arun Yun, Jihyun Fallone, B. Gino Univ Alberta Dept Oncol Med Phys Div 11560 Univ Ave Edmonton AB T6G 1Z2 Canada Cross Canc Inst Dept Med Phys 11560 Univ Ave Edmonton AB T6G 1Z2 Canada Cross Canc Inst Dept Radiat Oncol 11560 Univ Ave Edmonton AB T6G 1Z2 Canada Univ Alberta Dept Oncol Radiat Oncol Div 11560 Univ Ave Edmonton AB T6G 1Z2 Canada British Columbia Canc Victoria Radiat Oncol 2410 Lee Ave Victoria BC V8R 6V5 Canada Univ Calgary Tom Baker Canc Ctr Dept Oncol Radiat Oncol 1331 29th St NW Calgary AB T2N 4N2 Canada

Linear accelerator-magnetic resonance (linac-MR) hybrid systems allow for real-time magnetic resonance imaging (MRI)-guided radiotherapy for more accurate dose delivery to the tumor and improved sparing of the adjacent healthy tissues. However, for real-time tumor detection, it is unfeasible for a human expert to manually contour (gold standard) the tumor at the fast imaging rate of a linac-MR. This study aims to develop a neural network-based tumor autocontouring algorithm with patient-specific hyperparameter optimization (HPO) and to validate its contouring accuracy using in vivo MR images of cancer patients. Two-dimensional (2D) intrafractional MR images were acquired at 4 frames/s using 3 tesla (T) MRI from 11 liver, 24 prostate, and 12 lung cancer patients. A U-Net architecture was applied for tumor autocontouring and was further enhanced by implementing HPO using the Covariance Matrix Adaptation Evolution Strategy. Six hyperparameters were optimized for each patient, for which intrafractional images and experts' manual contours were input into the algorithm to find the optimal set of hyperparameters. For evaluation, Dice's coefficient (DC), centroid displacement (CD), and Hausdorff distance (HD) were computed between the manual contours and autocontours. The performance of the algorithm was benchmarked against two standardized autosegmentation methods: non-optimized U-Net and nnU-Net. For the proposed algorithm, the mean (standard deviation) DC, CD, and HD of the 47 patients were 0.92 (0.04), 1.35 (1.03), and 3.63 (2.17) mm, respectively. Compared to the two benchmarking autosegmentation methods, the proposed algorithm achieved the best overall performance in terms of contouring accuracy and speed. This work presents the first tumor autocontouring algorithm applicable to the intrafractional MR images of liver and prostate cancer patients for real-time tumor-tracked radiotherapy. The proposed algorithm performs patient-specific HPO, enabling accurate tumor deli

关键词： autocontouring intrafractional motion management tumor tracking linac-MR hybrid MRI guidance radiotherapy hyperparameter optimization

来源：评论

学校读者我要写书评

暂无评论

Surrogate-based automated hyperparameter optimization for expensive automotive crashworthiness optimization

引用

STRUCTURAL AND MULTIDISCIPLINARY optimization 2025年第4期68卷 1-19页

作者： Long, Fu Xing van Stein, Niki Frenzel, Moritz Krause, Peter Gitterle, Markus Back, Thomas BMW Grp Knorrstr 147 D-80788 Munich Germany Leiden Univ LIACS Niels Bohrweg 1 NL-2333 Leiden Netherlands Altair Engn GmbH Calwer Str 7 D-71034 Boblingen Germany Divis Intelligent Solut GmbH Joseph von Fraunhofer Str 20 D-44227 Dortmund Germany Munich Univ Appl Sci Dachauer Str 98b D-80335 Munich Germany

In the automotive industry, solving crashworthiness optimization problems efficiently is crucial to minimize time and cost investment on expensive function evaluations, e.g., using simulation runs. Nonetheless, automotive crashworthiness optimization is time-consuming and challenging even with domain knowledge, due to the fact that crash problems are typically high-dimensional, nonlinear, and discontinuous. In this work, we propose an automated hyperparameter optimization (HPO) approach for expensive black-box optimization (BBO) problems that can assist practitioners to solve automotive crash problems efficiently using optimally configured optimization algorithms. Precisely, the landscape characteristics of BBO problems, e.g., quantified using exploratory landscape analysis (ELA), are analyzed to identify cheap-to-evaluate representative functions that belong to the same optimization problem class. Based on these representative functions, algorithm configurations can be optimally fine-tuned at a relatively low computational cost. Using three optimization algorithms, consisting of modular covariance matrix adaptation evolutionary strategy (CMA-ES), modular differential evolution (DE), and Bayesian optimization (BO), we evaluate the potential of our approach based on the black-box optimization benchmarking (BBOB) suite and an automotive side crash problem. Since the optimal configurations identified using our approach can perform well on most of the BBOB functions, we believe that our approach can generalize well to BBO problems with similar optimization complexity. For the automotive side crash problem, the BO configuration fine-tuned using our approach can outperform the default BO configuration as well as the conventional response surface method (RSM), in terms of the best-found-solution and convergence speed. Furthermore, better solutions can be identified using the proposed approach compared to successive RSM (SRSM), when dealing with complex crash functions and

关键词： Black-box optimization hyperparameter optimization Representative functions Bayesian optimization Exploratory landscape analysis Automotive crashworthiness

来源：评论

学校读者我要写书评

暂无评论

Accelerating hyperparameter optimization with a secretary

引用

NEUROCOMPUTING 2025年 625卷

作者： Munoz, Victor Ballester, Carmen Copaci, Dorin Moreno, Luis Blanco, Dolores Univ Carlos III Madrid Syst Engn & Automat Dept Avda Univ 30 Leganes 28911 Madrid Spain

hyperparameter optimization (HPO) is a vital step in machine learning (ML) for enhancing model performance. However, the vast and complex nature of the search space makes HPO both challenging and resource- intensive. Automatic HPO methods have demonstrated their ability to efficiently explore high-dimensional hyperparameter spaces and identify optimal solutions, but training and evaluating the model for each set of hyperparameters is still computationally expensive. To further reduce the computational cost, we propose a novel strategy that wraps the HPO process and terminates it based on the sequence of hyperparameters evaluated. The algorithm is inspired by the classic secretary problem, with two additional variations to better adjust to the HPO process. We evaluated the algorithm using popular HPO samplers, including Random Search (RS), Tree-structured Parzen Estimator (TPE), Bayesian optimization with Gaussian Processes (BOGP), Genetic Algorithms (GA), and Particle Swarm optimization (PSO). Results indicate that the proposed algorithm accelerates the HPO process by an average of 34%, with only a minimal trade-off in solution quality of 8%. The algorithm is straightforward to implement, compatible with any HPO setup, and particularly effective in the early stages of optimization. This makes it a valuable tool for practitioners aiming to quickly identify promising hyperparameters or reducing the search space, significantly cutting down the time and computational resources required.

关键词： hyperparameter optimization Secretary problem Machine learning

来源：评论

学校读者我要写书评

暂无评论

Multi-objective hyperparameter optimization of artificial neural network in emulating building energy simulation

引用

ENERGY AND BUILDINGS 2025年 337卷

作者： Ibrahim, Mahdi Harkouss, Fatima Biwole, Pascal Fardoun, Farouk Ouldboukhitine, Salah-Eddine Univ Clermont Auvergne Inst Pascal CNRS SIGMA Clermont F-63000 Clermont Ferrand France Al Maaref Univ Fac Engn Beirut Lebanon PSL Res Univ Ctr Proc MINES Paris PERSEE CS 10207 F-06904 Sophia Antipolis France Calif State Polytech Univ Sch Engn Arcata CA 95521 USA Lebanese Univ Fac Engn Beirut Lebanon

Artificial Neural Networks (ANNs) play a significant role in emulating Building Energy Simulation (BES), forecasting building energy consumption, and optimizing energy retrofit measures. Determining the appropriate architecture for an ANN that can manage multiple predictions simultaneously is a complex task that requires extensive experimentation and validation to achieve optimal performance. To address this challenge, a novel methodology referred to as Multi-Objective hyperparameter optimization of ANN (MOHO-ANN) is introduced. This approach involves aligning ANN prediction results with data to achieve optimal performance by tuning the ANN's hyperparameters. The methodology consists of calibrating the BES model, creating data using a model sampling step for ANN training, and formulating multi-objective optimization using hyperparameter tuning to obtain the set of optimal ANN architectures. Lastly, a Multi-Criteria Decision-Making (MCDM) step is employed to select the optimal ANN. This method is applied to retrofit an existing building by incorporating weather data, passive, and renewable retrofit measures. The ANN is used to predict hourly energy consumption, energy generation, and thermal comfort in the retrofit scenario. 1.75 million hourly data points have been used to train, validate, and test the ANN model. The results underscore the practicality of MOHO-ANN procedure in achieving predictions, demonstrating a coefficient of determination (R2) exceeding 0.98. Further research directions should consider employing the optimal architectural model to perform retrofit optimization by incorporating climate change.

关键词： Artificial Neural Network Machine learning hyperparameter optimization Building energy simulation Multi criteria decision making Multi-objective optimization

来源：评论

学校读者我要写书评

暂无评论

A Novel hyperparameter optimization Approach for Supervised Classification: Phase Prediction of Multi-Principal Element Alloys

引用

COGNITIVE COMPUTATION 2025年第1期17卷 1-14页

作者： Fatimi, Syed Hassan Wang, Zidong Chang, Isaac T. H. Liu, Weibo Liu, Xiaohui Brunel Univ London Dept Comp Sci Uxbridge UB8 3PH Middlesex England Brunel Univ London Brunel Ctr Adv Solidificat Technol BCAST Uxbridge UB8 3PH Middlesex England

In this paper, a hyperparameter optimization approach is proposed for the phase prediction of multi-principal element alloys (MPEAs) through the introduction of two novel hyperparameters: outlier detection and feature subset selection. To gain a deeper understanding of the connection between alloy phases and their elemental properties, an artificial neural network is employed, with hyperparameter optimization performed using a genetic algorithm to select the optimum hyperparameters. The two novel hyperparameters, outlier detection and feature subset selection, are introduced within the optimization framework, along with new crossover and mutation operators for handling single and multi-valued genes simultaneously. Ablation studies are conducted, illustrating an improvement in prediction accuracy with the inclusion of these new hyperparameters. A comparison with five existing algorithms in multi-class classification is made, demonstrating an improvement in the performance of phase prediction, thereby providing a better perception of the alloy phase space for high-throughput MPEA design.

关键词： hyperparameter optimization Outlier detection Feature selection Artificial neural networks Genetic algorithm Multi-principal element alloys

来源：评论

学校读者我要写书评

暂无评论

Least Squares Support Vector Machines With Variable Selection and hyperparameter optimization for Complex Structures Reliability Assessment

引用

QUALITY AND RELIABILITY ENGINEERING INTERNATIONAL 2025年第4期41卷 1461-1470页

作者： Dong, Xiaowei Zhang, Hao Li, Zhenao Zhu, Chunyan Yi, Shujuan Chen, Changhai Heilongjiang Bayi Agr Univ Sch Engn Daqing Heilongjiang Peoples R China Minist Agr & Rural Affairs Inspect & Test Ctr Agr Prod & Processed Prod Qual Daqing Heilongjiang Peoples R China Heilongjiang Bayi Agr Univ Coll Elect & Informat Daqing Heilongjiang Peoples R China Harbin Acad Agr Sci Harbin Heilongjiang Peoples R China

For effectively estimating the reliability of complex structures, a least squares support vector machine with variable selection and hyperparameter optimization (SVMSO, short for) is proposed based on local linear embedding with Pearson coefficient and location density with particle swarm optimization (LDPSO) algorithm. In this proposed method, the local linear embedding with Pearson coefficient is used to select the variables that have a strong correlation with output responses, which are embedded in relatively low-dimensional space to avoid the negative influence of high-dimensional input parameters. The optimal hyperparameters of least squares support vector machines (LSSVM) are obtained by applying the LDPSO to improve the accuracy of LSSVM affected by the hyperparameters. Taking civil aircraft turbine blisk as a study case, the effectiveness and applicability of SVMSO are verified in aspects of modeling quality and simulation characteristics, by comparing direct simulation, support vector machine, and LSSVM. The case results and conclusions represent that the proposed method has good precision and efficiency under a high-dimensional data scale, and is suitable for reliability analysis of complex structures.

关键词： hyperparameter optimization least squares support vector machines reliability analysis turbine blisk variable selection

来源：评论

学校读者我要写书评

暂无评论

Deep neural network based distribution system state estimation using hyperparameter optimization

引用

RESULTS IN ENGINEERING 2024年 24卷

作者： Bekesi, Gergo Barancsuk, Lilla Hartmann, Balint Budapest Univ Technol & Econ Fac Elect Engn & Informat Dept Automat & Appl Informat Muegyetem Rkp 3 H-1111 Budapest Hungary Budapest Univ Technol & Econ Fac Elect Engn & Informat Dept Elect Power Engn Muegyetem Rkp 3 H-1111 Budapest Hungary

In the past decade, distribution system state estimation has become a crucial topic in power system research due to the increasing importance of distribution networks amidst the decline of centralized energy production. This paper addresses a gap in the literature regarding the application of modern hyperparameter optimization techniques in low-voltage distribution system state estimation using deep neural networks. In particular, it demonstrates the use of the Tree-structured Parzen Estimator algorithm, which is a Bayesian hyperparameter optimization method, for distribution system state estimation on real Hungarian low-voltage networks. The study uses data from four real-life low-voltage supply areas in Hungary, which were modeled to address the challenges in obtaining network information. Compared to traditional methods like the weighted least squares method, the Tree-structured Parzen Estimator algorithm significantly improves the accuracy of the voltage amplitude and angle estimations, reducing the relative error by 14-73%. Additionally, it is shown that TPE outperforms simpler methods like Random Search in hyperparameter optimization. The results also reveal connections between the distribution system size and optimal hyperparameters, such as batch size, learning rate, and hidden layer configuration. The proposed non- iterative algorithm, combined with the parallel computation capabilities of deep neural networks utilizing GPU, resulted in four orders of magnitude improvement in runtime. These advancements make the proposed approach a valuable tool for renewable energy integration planning and real-time monitoring, highlighting its potential for practical applications in the power industry.

关键词： Deep neural network Distribution system state estimation hyperparameter optimization Low-voltage Tree-structured Parzen Estimator

来源：评论

学校读者我要写书评

暂无评论

Sound absorption performance prediction of multi-dimensional Helmholtz resonators based on deep learning and hyperparameter optimization

引用

PHYSICA SCRIPTA 2025年第2期100卷 026013-026013页

作者： Liu, Yan Hang, Yin Li, Qiutong Shanghai Res Inst Mat Co Ltd Acoust Metamat Lab Shanghai 200437 Peoples R China Shanghai Res Inst Mat Ltd Shanghai Key Lab Engn Mat Applicat & Evaluat Shanghai 200437 Peoples R China Shanghai Polytech Univ Sch Energy & Mat Shanghai 201209 Peoples R China

The problem of low-frequency noise is becoming increasingly severe and measuring the sound absorption performance of acoustic metamaterials (AMs) using accurate absorption coefficients is of great interest in low-frequency noise control engineering. Conventional calculation methods such as Finite Element Method (FEM) simulations and the theoretical analysis methods (TAM) have specific limitations. Deep learning (DL) models provide new perspectives for studying AMs acoustic performance. However, the prediction performance of DL models is highly dependent on the proper tuning of hyperparameters. As far as is known, existing literature has not systematically explored the impact of hyperparameter tuning on DL models in the context of acoustic performance studies. The present paper designed a multi-dimensional Helmholtz resonator (MDHR) consisting of a 4 x 4-type continuous parallel arrangement, while a dataset was established via FEM. Furthermore, a deep neural network (HPO-DNNs) model based on hyperparameter optimization (HPO) was proposed to predict the acoustic performance of the MDHR. Random search (RS), Bayesian optimization (BO), Simulated annealing (SA), and genetic algorithm (GA) were introduced to optimize the hyperparameters (learning rate, weight decay, optimizer, and batch size) of the DNNs. The mean square error (MSE), coefficient of determination (R2) of the testing dataset and the optimization time were used as the evaluation metrics, GA was selected for further study based on the comparison results (MSE = 0.00177, R2 = 0.98151) of the optimization efficiency and predictive precision of DNNs from the four HPO algorithms. Finally, the prediction performance of the GA-DNNs model was evaluated in single-, multi-, and broadband conditions in practical applications, demonstrating high precision and stability and providing a new approach for acoustics performance studies.

关键词： helmholtz resonator deep learning hyperparameter optimization

来源：评论

学校读者我要写书评

暂无评论

An improved decision tree model through hyperparameter optimization using a modified gray wolf optimization for diabetes classification

引用

COMPUTER METHODS IN BIOMECHANICS AND BIOMEDICAL ENGINEERING 2025年 1-17页

作者： Sam'an, Muhammad Farikhin, Muhammad Munsarif, Muhammad Univ Muhammadiyah Semarang Dept Informat Semarang Indonesia Univ Muhammadiyah Malaysia Fac Business Management & Informat Technol Semarang Malaysia Univ Diponegoro Dept Math Semarang Indonesia

Diabetes is a chronic condition that affects blood sugar levels and vital organs in the body. Early detection is crucial given the increasing global prevalence of diabetes and the grave risk of complications if not properly managed. Thus, a good prediction system is necessary. Although the Decision Tree (DT) is commonly used for classification, it is less effective with large datasets. We propose hyperparameter optimization of the DT using the Grey Wolf optimization (GWO), which has exploration and both exploitation capabilities. However, the limited search space of GWO may hinder practical exploration and exploitation, leading to premature optimization. To address this, we propose a modified GWO (MGWO) by adding the Levy distribution function to enhance the movements of alpha, beta, and delta wolves. We also provide GA (Genetic Algorithm) as a comparative algorithm. The fitness value of MGWO is 0.8498, surpassing GWO (0.8373) and GA (0.8492). Evaluation results indicate that MGWO and GA yield similar and superior accuracy compared to GWO. The proposed method outperforms existing ones. Further research is needed to evaluate the impact of varying the number of wolves on optimization performance and classification accuracy.

关键词： Diabetes decision tree hyperparameter optimization grey wolf optimization classification

来源：评论

学校读者我要写书评

暂无评论

MetaGen: A framework for metaheuristic development and hyperparameter optimization in machine and deep learning

引用

NEUROCOMPUTING 2025年 637卷

作者： Gutierrez-aviles, David Jimenez-navarro, Manuel Jesus Torres, Jose Francisco Martinez-Alvarez, Francisco Univ Seville Dept Comp Sci ES-41012 Seville Spain Pablo de Olavide Univ Data Sci & Big Data Lab ES-41013 Seville Spain

hyperparameter optimization is a pivotal step in enhancing model performance within machine learning. Traditionally, this challenge is addressed through metaheuristics, which efficiently explore large search spaces to uncover optimal solutions. However, implementing these techniques can be complex without adequate development tools, which is the primary focus of this paper. Hence, we introduce MetaGen, a novel Python package designed to provide a comprehensive framework for developing and evaluating metaheuristic algorithms. MetaGen follows best practices in Python design, ensuring minimalistic code implementation, intuitive comprehension, and full flexibility in solution representation. The package defines two distinct user roles: Developers, responsible for algorithm implementation for hyperparameter optimization, and Solvers, who leverage pre-implemented metaheuristics to address optimization problems. Beyond algorithm implementation, MetaGen facilitates benchmarking through built-in test functions, ensuring standardized performance comparisons. It also provides automated reporting and visualization tools to analyze optimization progress and outcomes effectively. Furthermore, its modular design allows distribution and integration into existing machine learning workflows. Several illustrative use cases are presented to demonstrate its adaptability and efficacy. The package, along with code, a user manual, and supplementary materials, is available at: https://***/Data-Science-Big-Data-Research-Lab/MetaGen.

关键词： Metaheuristics hyperparameter optimization Machine learning Deep learning Python

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：