检索结果-内蒙古大学图书馆

data-driven optimization of Complex Systems (DOCS), International Conference on

作者： Jingchuan Zhang Yaxue Liu Zhiming Dong Xianpeng Wang National Frontiers Science Center for Industrial Intelligence and Systems Optimization Northeastern University Shenyang China Key Laboratory of Data Analytics and Optimization for Smart Industry (Northeastern University) Ministry of Education Shenyang China

ISBN: (数字)9798350377842

ISBN: (纸本)9798350377859

In the process of multi-objective optimization, a large amount of data is generated, which contains the problem-related information. Learning knowledge from this data and applying it to the evolutionary process can improve the quality of the offspring solutions. Most existing knowledge learning methods suffer from high algorithm complexity, and there are still few methods designed specifically for the decomposition-based MOEAs. This paper proposes an evolutionary direction vector (EDV)-guided multiobjective evolutionary algorithm based on decomposition (MOEA/D-EDV). It collects the high-quality evolutionary process data based on the solution's performance in the objective space and their movement directions in the decision space. It then extracts the problem-related knowledge using an efficient learning method to accelerate the evolutionary process. Experimental results on 30 benchmark problems demonstrate that the EDV-guided evolution can significantly accelerate the evolutionary process, and the overall performance of the proposed MOEA/D-EDV is significantly better than the comparative algorithms in the literature.

关键词： Learning systems Evolutionary computation Benchmark testing Vectors Complexity theory Complex systems optimization

来源：评论

学校读者我要写书评

暂无评论

Evolution and detection of vector superradiant instabilities

引用

Physical Review D 2024年第8期110卷 083029页

作者： Yin-Da Guo Nayun Jia Shou-Shan Bao Hong Zhang Xin Zhang Key Laboratory of Particle Physics and Particle Irradiation (Ministry of Education) Institute of Frontier and Interdisciplinary Science Department of Physics Key Laboratory of Cosmology and Astrophysics (Liaoning) and College of Sciences Key Laboratory of Data Analytics and Optimization for Smart Industry (Ministry of Education) National Frontiers Science Center for Industrial Intelligence and Systems Optimization

Ultralight vectors can extract energy and angular momentum from a Kerr black hole (BH) due to superradiant instability, resulting in the formation of a BH-condensate system. In this work, we carefully investigate the evolution of this system numerically with multiple superradiant modes. Simple formulas are obtained to estimate important timescales, maximum masses of different modes, as well as the BH mass and spin at various times. Due to the coexistence of modes with small frequency differences, the BH-condensate system emits gravitational waves with a unique beat signature, which could be directly observed by current and projected interferometers. Besides, the current BH spin-mass data from the binary BH merger events already exclude the vector mass in the range 5×10−15 eV<μ<9×10−12 eV.

关键词： Gravitational wave sources Phenomenology Vector mesons

来源：评论

学校读者我要写书评

暂无评论

A Method for Predicting Mechanical Properties of Steel Based on Graph Convolutional Networks

SSRN

引用

SSRN 2024年

作者： Ban, Yunqi Yang, Yang Wang, Xianpeng Wu, Zhiyuan National Frontiers Science Center for Industrial Intelligence and Systems Optimization Northeastern University Shenyang110819 China Key Laboratory of Data Analytics and Optimization for Smart Industry Northeastern University Ministry of Education Shenyang110819 China Liaoning Engineering Laboratory of Data Analytics and Optimization for Smart Industry Shenyang China College of Electronic and Information Engineering Tongji University Shanghai201804 China Northeastern University China

In steel materials, several mechanical properties are interrelated and there is an inherent topology structure among these mechanical properties. Based on this motivation, a multitask neural network model based graph convolutional network is proposed to predict mechanical properties. First, a directed graph is built over the mechanical properties, with each mechanical property representing a node on the graph. Then this graph convolutional network, which represents the information propagation among mechanical properties, is combined with a multitask neural network and trained together. By comparing with K-NearestNeighbor (KNN), support vector regression (SVR), random forest (RF), Extreme Gradient Boosting (XGBoost), Cross-Stitch Network, Multi-gate Mixture-of-Experts (MMOE) and multitask neural (MTN) algorithms, the proposed algorithm is optimal in all of the metrics root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R2) that indicate the predictive performance on the prediction of tensile strength (TS), yield strength (YS) and elongation (EL). Specifically, R2 is 0.9797 in tensile strength, R2 is 0.9747 in yield strength and R2 is 0.8316 in elongation. In addition, a deep representation of the various influences on mechanical properties is obtained through the parameters of the model with validity and interpretability. Material insights are extracted from this deep representation, including the comprehensive interaction of processing and chemical composition on the mechanical properties as well as the mechanical properties with each other. Further, the prediction accuracy can be improved by the combination of different structures of neural networks with graph convolutional network to prove the effectiveness of graph convolutional network. © 2024, The Authors. All rights reserved.

关键词： Tensile strength

来源：评论

学校读者我要写书评

暂无评论

Metavit-Trans: A Framework for Mixed-Type Defect Detection of Wafers with Vision Transformer Combined with Meta-Learning and Transfer Learning

Metavit-Trans: A Framework for Mixed-Type Defect Detection o...

引用

China Semiconductor Technology International Conference (CSTIC)

作者： Junfeng Zhao Lixin Tang National Frontiers Science Center for Industrial Intelligence and Systems Optimization Northeastern University Shenyang China Ministry of Education Key Laboratory of Data Analytics and Optimization for Smart Industry (Northeastern University) Shenyang China

The defect detection of semiconductor wafer patterns is essential in the process of chip manufacturing. With the improvement of the IC process and design level, the types of wafer surface defects become more complex. In addition, the sample size of wafer map defect data is usually small, and the class is unbalanced, which poses higher challenges to the existing methods. To effectively identify mixed-type defects, a mixed-type defect detection framework with Vision Transformer (ViT) combined with meta-learning and transfer learning (MetaViT-Trans) is proposed. The results show that the MetaViT-Trans framework has a good defect feature extraction ability for imbalanced small-scale wafer map defect data, and has a good multi-scale information learning ability, which can effectively identify a variety of mixed-type defects.

关键词：

来源：评论

学校读者我要写书评

暂无评论

MOEA/D With Spatial-Temporal Topological Tensor Prediction for Evolutionary Dynamic Multiobjective optimization

引用

IEEE Transactions on Evolutionary Computation 2024年 1-1页

作者： Wang, Xianpeng Zhao, Yumeng Tang, Lixin Yao, Xin Ministry of Education Shenyang China Liaoning Engineering Laboratory of Data Analytics and Optimization for Smart Industry Liaoning Key Laboratory of Manufacturing System and Logistics Optimization Shenyang China National Frontiers Science Center for Industrial Intelligence and Systems Optimization Shenyang China Lingnan University Hong Kong China

When solving dynamic multiobjective optimization problems, most evolutionary algorithms attempt to predict the initial population in a new environment by mining the relationships between solutions during historical environment changes. However, the complex relationships between solutions and the limited amount of available data often make it difficult to extract useful information efficiently, which may deteriorate the prediction accuracy. To address this problem, this paper proposes a spatial-temporal topological tensor-based prediction method to generate the initial population in a new environment under the decomposition framework of MOEA/D. The method relies on the idea that the population distribution in each environment has topological similarity along the time dimension in the objective space, which makes it efficient to represent the population distribution in terms of a tensor and predict new solutions along each decomposition axis in a new environment by an improved tensor-based multi-short time series prediction method. Experimental results on various benchmark problems and a real-world problem show that the proposed method is competitive or even superior to state-of-the-art dynamic multiobjective evolutionary algorithms based on prediction strategies. IEEE

关键词： Tensors

来源：评论

学校读者我要写书评

暂无评论

Lithography Hotspot Detection Based on Transfer Learning With High Resolution Networks

Lithography Hotspot Detection Based on Transfer Learning Wit...

引用

China Semiconductor Technology International Conference (CSTIC)

作者： Hongzhe Wang Lixin Tang National Frontiers Science Center for Industrial Intelligence and Systems Optimization Northeastern University Shenyang China The Key Laboratory of Data Analytics and Optimization for Smart Industry (Northeastern University) Ministry of Education Shenyang China

As integrated circuit (IC) technology continues to advance, lithography hotspot detection is of importance in physical verification flow and can affect the turn-around time and the yield of IC manufacturing. In this paper, a deep learning-based lithography hotspot detection method is proposed to overcome the problem of unbalanced positive and negative samples in detection. The proposed method uses the High-Resolution Network (HR-Net18) and pre-trained model from ImageNet to improve transferability. ICCAD 2012 benchmark suits is used for model train and test. The experimental results show that the proposed method performs well in terms of the values of AUC and precision.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Prospects for measuring dark energy with 21 cm intensity mapping experiments

arXiv

引用

arXiv 2021年

作者： Wu, Peng-Ju Zhang, Xin Department of Physics College of Sciences Northeastern University Shenyang110819 China MOE Key Laboratory of Data Analytics and Optimization for Smart Industry Northeastern University Shenyang110819 China

Using the 21 cm intensity mapping (IM) technique can efficiently perform large-scale neutral hydrogen (H i) surveys, and this method has great potential for measuring dark-energy parameters. Some 21 cm IM experiments aiming at measuring dark energy in the redshift range of 0 0) = 0.080 and σ(wa) = 0.25 in the CPL model. Compared with CMB+BAO+SN, Tianlai can provide tighter constraints in ΛCDM and wCDM, but looser constraints (tighter than CMB+BAO) in CPL, and the combination CMB+BAO+SN+Tianlai gives σ(w) = 0.013, σ(w0) = 0.055, and σ(wa) = 0.13. In addition, it is found that the synergy of FAST (0 © 2021, CC BY.

关键词： Surveys

来源：评论

学校读者我要写书评

暂无评论

How can gravitational-wave standard sirens and 21-cm intensity mapping jointly provide a precise late-universe cosmological probe?

arXiv

引用

arXiv 2021年

作者： Jin, Shang-Jie Wang, Ling-Feng Wu, Peng-Ju Zhang, Jing-Fei Zhang, Xin Department of Physics College of Sciences Northeastern University Shenyang110819 China MOE Key Laboratory of Data Analytics and Optimization for Smart Industry Northeastern University Shenyang110819 China

In the next decades, the gravitational-wave (GW) standard siren observations and the neutral hydrogen 21-cm intensity mapping (IM) surveys, as two promising cosmological probes, will play an important role in precisely measuring cosmological parameters. In this work, we make a forecast for cosmological parameter estimation with the synergy between the GW standard siren observations and the 21-cm IM surveys. We choose the Einstein Telescope (ET) and the Taiji observatory as the representatives of the GW detection projects and choose the Square Kilometre Array (SKA) phase I mid-frequency array as the representative of the 21-cm IM experiments. In the simulation of the 21-cm IM data, we assume perfect foreground removal and calibration. We find that the synergy of the GW standard siren observations and the 21-cm IM survey could break the cosmological parameter degeneracies. The joint ET+Taiji+SKA data give σ(H0) = 0.28 km s−1 Mpc−1 in the ΛCDM model, σ(w) = 0.028 in the wCDM model, which are better than the results of Planck+BAO+SNe, and σ(w0) = 0.077 and σ(wa) = 0.295 in the CPL model, which are comparable with the results of Planck+BAO+SNe. In the ΛCDM model, the constraint precision of H0 and Ωm is less than or rather close to 1%, indicating that the magnificent prospects for precision cosmology with these two promising cosmological probes are worth expecting. © 2021, CC BY.

关键词： Surveys

来源：评论

学校读者我要写书评

暂无评论

Event-triggered control for switched descriptor systems based on sampled-data implementation

Event-triggered control for switched descriptor systems base...

引用

第33届中国控制与决策会议

作者： Jiasheng Shi Jun Zhao State Key Laboratory of Synthetical Automation for Process Industries Northeastern University Key Laboratory of Data Analytics and Optimization for Smart Industry (Northeastern University) Ministry of Education College of Information Science and Engineering Northeastern University

This paper studies event-triggered control problem for a class of switched descriptor systems based on sampled-data *** switched descriptor systems are first to be decomposed into a equivalent form,and then a reduced-order switched system is ***,a sufficient condition ensures that the closed-loop system is exponentially stable under average dwell time switching ***,the Zeno behavior is ruled *** the end,a numerical examples is demonstrated the merit and effectiveness of the proposed methods.

关键词： Average dwell time Switched descriptor systems Event-triggered control Sampled-data

来源：评论

学校读者我要写书评

暂无评论

Finite-time H∞ Control for Switched Systems:An Event-triggered Bumpless Transfer Approach

Finite-time H∞ Control for Switched Systems:An Event-trigge...

引用

第35届中国控制与决策会议

作者： Jinghan Li Guangxu He Qiang Zeng Jun Zhao State Key Laboratory of Synthetical Automation for Process Industries Northeastern University Key Laboratory of Data Analytics and Optimization for Smart Industry(Northeastern University) Ministry of Education College of Information Science and Engineering Northeastern University

In this paper,the event-triggered finite-time H∞bumpless transfer control problem of switched systems is *** event-triggered bumpless transfer control method with local constraints is proposed to reduce the bump generated by switching event-triggered controller.A sufficient condition is given to ensure the solvability of the switched systems with eventtriggered finite-time H bumpless transfer *** positive lower bound of event-triggered integral is obtained,which implies that Zeno behavior is *** using the multiple Lyapunov functions method,a state-dependent switching law,an eventtriggered mechanism and a bumpless transfer controller are ***,a simulation example is given to verify the effectiveness of the proposed method.

关键词： switched systems bumpless transfer event-triggered finite-time H_∞control

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：