检索结果-内蒙古大学图书馆

41st International Conference on Machine Learning (ICML)

作者： Bassily, Raef Cortes, Corinna Mao, Anqi Mohri, Mehryar Ohio State Univ Columbus OH USA Google Res New York NY USA Courant Inst Math Sci New York NY 10012 USA

In many applications, the labeled data at the learner's disposal is subject to privacy constraints and is relatively limited. To derive a more accurate predictor for the target domain, it is often beneficial to leverage publicly available labeled data from an alternative domain, somewhat close to the target domain. This is the modern problem of supervised domain adaptation from a public source to a private target domain. We present two ( epsilon,delta)-differentially private adaptation algorithms for supervised adaptation, for which we make use of a general optimization problem, recently shown to benefit from favorable theoretical learning guarantees. Our first algorithm is designed for regression with linear predictors and shown to solve a convex optimization problem. Our second algorithm is a more general solution for loss functions that may be non-convex but Lipschitz and smooth. While our main objective is a theoretical analysis, we also report the results of several experiments. We first show that the non-private versions of our algorithms match state-of-the-art performance in supervised adaptation and that for larger values of the target sample size or epsilon, the performance of our private algorithms remains close to that of their non-private counterparts.

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Heterogeneous Distributed Subgradient

Heterogeneous Distributed Subgradient

引用

American Control Conference (ACC)

作者： Lin, Yixuan Gamarra, Marco Liu, Ji Meta Menlo Pk CA USA SUNY Stony Brook Dept Appl Math & Stat Stony Brook NY USA Air Force Res Lab Wright Patterson AFB OH 45433 USA SUNY Stony Brook Dept Elect & Comp Engn Stony Brook NY USA

ISBN: (纸本)9798350382662;9798350382655

The paper proposes a heterogeneous push-sum based subgradient algorithm for multi-agent distributed convex optimization, in which each agent can arbitrarily switch between subgradient-push and push-subgradient at any time. It is shown that the heterogeneous algorithm converges to an optimal point at an optimal rate over time-varying directed graphs. The switching process within the heterogeneous algorithm can help prevent the leakage of agents' subgradient information.

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

COVID-HGS: Early severity prediction model of COVID-19 based on Hunger Game Search (HGS) optimization algorithm

引用

BIOMEDICAL SIGNAL PROCESSING AND CONTROL 2024年第PartA期96卷

作者： Sayed, Safynaz AbdEl-Fattah Elkorany, Abeer Sayed, Sabah Beni Suef Univ Fac Comp & Artificial Intelligence Dept Comp Sci Bani Suwayf Egypt Cairo Univ Fac Comp & Artificial Intelligence Dept Comp Sci Cairo Egypt

COVID-19 is a rapidly spreading global pandemic that causes lung problems and a lack of oxygen in the blood, which leads to death. The emergence of this disease caused a shortage of hospital resources due to a failure to prioritize patients correctly according to their severity, which led to a waste of resources. Currently, X-ray images play a vital role in early detection and monitoring the different states of COVID-19 patients. Every patient may have many X-ray images at different times to follow up on his health and needs, so detecting the most suitable stage of patient images for predicting his severity and needs is urgent to save his life and hospital resources. This paper proposes a prediction model to early predict three levels of severity risks for the COVID-19 patient based on X-ray images: high, moderate, and low. Pre-processing, extracting, and selecting the features based on X-ray images are critical and effective phases. Thus, in the proposed model, a combination of methods are used for the pre-processing phase, different handcrafted and deep learning techniques are introduced for the feature extraction step, and a merged meta-heuristic optimization algorithm called Hunger Game Search (HGS) with support vector machine (SVM) is proposed to automatically determine the most relevant features for the prediction model. The experimental results demonstrated that using the processed images is better than using the original images, where the pre-trained deep learning model Visual Geometry Group (VGG-19) has achieved the best results in the feature extraction phase compared with other pre-trained deep models and handcrafted methods, and the HGS optimization algorithm has accomplished the perfect results for selecting the features against other six popular meta-heuristic algorithms. The model achieved 99.9% and 94.8% for the best and average accuracy, respectively, at an early stage with the least number of selected features compared with the competitor meta-heur

关键词： Mortality and severity prediction optimization algorithms Feature selection Metaheuristic swarm intelligence COVID-19 X-ray images Hunger Game Search (HGS) algorithm

来源：评论

学校读者我要写书评

暂无评论

Hybridization of Moth flame optimization algorithm and quantum computing for gene selection in microarray data

引用

JOURNAL OF AMBIENT INTELLIGENCE AND HUMANIZED COMPUTING 2021年第2期12卷 2731-2750页

作者： Dabba, Ali Tari, Abdelkamel Meftali, Samy Mohamed Boudiaf Univ Fac Math & Comp Sci Comp Sci Dept Msila Algeria Abderrahmane Mira Univ Fac Sci Comp Sci Dept Bejaia Algeria Med Comp Lab LIMED Bejaia Algeria Univ Lille Lille France Res Ctr Comp Sci Signal & Automat Control Lille C Lille France

Ever-increasing data in various fields like Bioinformatics field, which has led to the need to find a way to reduce the data dimensionality. Gene selection problem has a large number of genes (relevant, redundant or noise), which needs an effective method to help us in detecting diseases and cancer. In this problem, computational complexity is reduced by selecting a small number of genes, but it is necessary to choose the relevant genes to keep a high level of accuracy. Therefore, in order to find the optimal gene subset, it is essential to devise an effective exploration approach that can investigate a large number of possible gene subsets. In addition, it is required to use a powerful evaluation method to evaluate the relevance of these gene subsets. In this paper, we present a novel swarm intelligence algorithm for gene selection called quantum moth flame optimization algorithm (QMFOA), which based on hybridization between quantum computation and moth flame optimization (MFO) algorithm. The purpose of QMFOA is to identify a small gene subset that can be used to classify samples with high accuracy. The QMFOA has a simple two-phase approach, the first phase is a pre-processing that uses to address the difficulty of high-dimensional data, which measure the redundancy and the relevance of the gene, in order to obtain the relevant gene set. The second phase is a hybridization among MFOA, quantum computing, and support vector machine with leave-one-out cross-validation, etc., in order to solve the gene selection problem. We use quantum computing to guarantee a good trade-off between the exploration and the exploitation of the search space, while a new update moth operation using Hamming distance and Archimedes spiral allows an efficient exploration of all possible gene-subsets. The main objective of the second phase is to determine the best relevant gene subset of all genes obtained in the first phase. In order to assess the performance of the proposed QMFOA, we test Q

关键词： Gene expression Feature selection Moth flame optimization algorithm Quantum computing Microarray data Cancer classification Bio-inspired algorithms Molecular biology optimization algorithms Evolutionary algorithms Swarm intelligence

来源：评论

学校读者我要写书评

暂无评论

Application of SAM-based prompting algorithm in camouflage object detection 5

Application of SAM-based prompting algorithm in camouflage o...

引用

5th International Conference on Computer Engineering and Application (ICCEA)

作者： Yuan, Jiatang Hubei Univ Technol Sch Comp Sci Wuhan Peoples R China

ISBN: (纸本)9798350386783;9798350386776

We propose a prompting algorithm based on SAM and apply it to the field of camouflage object detection. We extract the camouflaged object detection (COD) network features after weakly supervised learning and generate bounding box information and point prompt information to guide the SAM to segment the camouflaged objects in the image. The proposed method can effectively improve the performance of camouflage object detection network and has achieved good results on the mainstream camouflage object detection dataset.

关键词： SAM Camouflage object detection optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Fast NMPC Design for Image-Based Visual Servoing of Autonomous Underwater Vehicles 7

Fast NMPC Design for Image-Based Visual Servoing of Autonomo...

引用

IEEE 7th International Conference on Industrial Cyber-Physical Systems (ICPS)

作者： Gu, Hang Shen, Chao Carleton Univ Dept Syst & Comp Engn Ottawa ON Canada

ISBN: (纸本)9798350363029;9798350363012

This paper presents a new approach to image-based visual servoing (IBVS) for Autonomous Underwater Vehicles (AUVs) with the goal of improved performance and computational efficiency. Traditional IBVS methods, when combined with Model Predictive Control (MPC), face high computational demands due to the nonlinear dynamics and the large degrees of freedom (DOFs) in the variables of the associated optimization problem. Our method addresses this by reducing the DOFs of the optimization variable in the cost function while maintaining a good control performance. To further consider the smoothness of the MPC control signal, a soft constraint handling method is developed. The fast nonlinear MPC, combined with smoother control trajectories and effective constraint handling, makes our method particularly suitable for AUV IBVS applications in dynamic environments. Comparisons with standard strategies confirm the improved performance of our approach in terms of both speed and trajectory quality. Simulation results show that our approach can achieve an improved computation up to 100 times faster than conventional MPC-based IBVS methods, which highlights the great potential for real-time IBVS applications.

关键词： Model Predictive Control Visual Servoing Computational Efficiency Robotics optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Load Frequency Control in Two Area Power Systems in A Smart Grid Environment 12

Load Frequency Control in Two Area Power Systems in A Smart ...

引用

12th International Conference on Smart Grid (ICSmartGrid)

作者： Mohamed, Faradji Layadi, Toufik Madani Colak, Ilhami Univ MHamed Bougara Boumerdes Lab Syst Engn & Telecommun Boumerdes Algeria Univ Mohamed El Bachir El Ibrahimi Labof Mat & Elect Syst Bordj Bou Arreridj Algeria Istinye Univ Dept Elect Engn Istanbul Turkiye

ISBN: (纸本)9798350361612;9798350361629

Load Frequency Control (LFC) is a critical aspect of power system stability, ensuring that the frequency and tie-line power flow remains within acceptable limits. In this paper, we investigate LFC in a two-area system with the integration of demand response (DR) loops. The DR loops allow for dynamic adjustments of load demand based on real-time system conditions. Our study focuses on optimizing the proportional-integral-derivative (PID) controller used in the LFC system. To achieve this, we perform a comparative analysis of three optimization algorithms: Artificial bee colony (ABC), particle swarm optimization (PSO), and Aquila optimization (AO). These algorithms are applied to tune the PID controller parameters, aiming to enhance system performance, reduce frequency deviations, and minimize control efforts. Simulation results demonstrate the effectiveness of the proposed approach. The optimized PID controller, combined with DR, significantly improves system response during load disturbances. Furthermore, the comparative study sheds light on the strengths and weaknesses of each optimization algorithm, providing valuable insights for future LFC implementations. Overall, our work contributes to the advancement of LFC strategies in interconnected power systems, emphasizing the role of demand response and optimization techniques in achieving robust and efficient control

关键词： Load Frequency Control demand response style optimization algorithms proportional-integral-derivative

来源：评论

学校读者我要写书评

暂无评论

Supercritical thermophysical properties prediction of multi-component hydrocarbon fuels based on artificial neural network models

引用

Science China(Technological Sciences) 2022年第4期65卷 903-919页

作者： MING ZhenYang LIU HaiFeng WANG QianLong YUE ZongYu CUI YanQing WEN MingSheng YAO MingFa State Key Laboratory of Engines Tianjin UniversityTianjin 300072China

A good understanding of the thermophysical properties of hydrocarbon fuels at supercritical pressure is important to research on experiment and numerical simulation of fuel supercritical *** measurements are difficult to conduct directly because of the extremely high pressure and high *** this study,back propagation(BP)neural network,BP optimized by mind evolution algorithm(MEA-BP)and BP neural network optimized by genetic algorithm(GA-BP)are established to determine the nonlinear temperature-dependent thermophysical properties of density,viscosity,and isobaric specific heat(C_(2))of hydrocarbon fuels at supercritical ***,approximate formulas for these properties prediction are primarily proposed using polynomial *** this paper,models that can predict three types of physical properties of three kinds of hydrocarbon fuels and their mixtures in a wide temperature range under supercritical pressure are *** the prediction of density and C_(2),BP neural network has a good prediction *** results show that the MAPE is lower than 2%in the prediction of density and C_(2),but the MAPE of viscosity prediction is slightly higher than 5%using ***,MEA and GA are used to optimize the prediction of *** optimization effect and computation of the MEA is better than that of GA because MEA does not have the local optimization and prematurity *** present work offers an efficient tool to predict the thermophysical properties of hydrocarbon fuels over a wide range of temperatures under supercritical pressure which can be easily extended to other fuels of *** will be beneficial to the experiment and numerical simulation studies of supercritical sprays.

关键词： alternative fuel supercritical spray back propagation neural network optimization algorithms thermophysical properties

来源：评论

学校读者我要写书评

暂无评论

Efficient mask optimization for enhanced digital maskless lithography quality by improved particle swarm optimization algorithm

JOURNAL OF VACUUM SCIENCE & TECHNOLOGY B

引用

JOURNAL OF VACUUM SCIENCE & TECHNOLOGY B 2024年第5期42卷

作者： Huang, Shengzhou Wu, Dongjie Tang, Yuanzhuo Ren, Bowen Pan, Jiani Tian, Zhaowei Li, Zhi Huang, Jinjin Anhui Polytech Univ Sch Artificial Intelligence Wuhu 241000 Peoples R China Anhui Polytech Univ Sch Mech Engn Wuhu 241000 Peoples R China Anhui East China Photoelect Technol Study Inst Wuhu 241002 Peoples R China Hefei Acer Microelect Equipment Co Ltd Hefei 230000 Peoples R China Wuhu Changpeng Auto Parts Co LTD Wuhu Peoples R China

In this paper, an efficient mask optimization method for enhanced digital micromirror device lithography quality based on improved particle swarm optimization (PSO) is proposed, which greatly improves the quality of lithography. First, the traditional PSO algorithm is improved by introducing adaptive parameter adjustment to enhance its search ability in complex problems. In addition, in order to avoid premature convergence of the algorithm, a simulated annealing operation is introduced to make it accept the different solution with a certain probability and jump out of the local optimal better. The numerical simulation experiment results showed that the pattern errors between the print image and target pattern were reduced by 93.5%, 95.8%, and 95.6%, respectively. Compared with traditional optimization methods, the proposed algorithm significantly improves the image quality, especially in the aspects of edge contour and pattern fidelity.

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Role of gaps in the digitized counterdiabatic quantum approximate optimization algorithm for fully connected spin models

引用

Physical Review A 2024年第6期110卷 062606-062606页

作者： Mara Vizzuso Gianluca Passarelli Giovanni Cantele Procolo Lucignano Dipartimento di Fisica “E. Pancini” Università degli Studi di Napoli “Federico II” Complesso Universitario M. S. Angelo via Cintia 21 80126 Napoli Italy CNR-SPIN c/o Complesso Universitario M. S. Angelo via Cintia 21 80126 Napoli Italy

Recently, digitized-counterdiabatic (CD) corrections to the quantum approximate optimization algorithm (QAOA) have been proposed, yielding faster convergence within the desired accuracy than standard QAOA. In this manuscript, we apply this approach to a fully connected spin model with random couplings. We show that the performances of the algorithm are related to the spectral properties of the instances analyzed. In particular, the larger the gap between the ground state and the first excited states, the better the convergence to the exact solution.

关键词： optimization algorithms

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：