检索结果-内蒙古大学图书馆

adaptive optimization for prediction with missing data

MACHINE LEARNING 2025年第5期114卷 1-37页

作者： Bertsimas, Dimitris Delarue, Arthur Pauphilet, Jean MIT Sloan Sch Management 77 Massachusetts Ave Cambridge MA USA Georgia Inst Technol H Milton Stewart Sch Ind & Syst Engn 755 Ferst Dr Atlanta GA 30332 USA London Business Sch Regents Pk London NW1 4SA England

When training predictive models on data with missing entries, the most widely used and versatile approach is a pipeline technique where we first impute missing entries and then compute predictions. In this paper, we view prediction with missing data as a two-stage adaptive optimization problem and propose a new class of models, adaptive linear regression models, where the regression coefficients adapt to the set of observed features. We show that some adaptive linear regression models are equivalent to learning an imputation rule and a downstream linear regression model simultaneously instead of sequentially. We leverage this joint-impute-then-regress interpretation to generalize our framework to non-linear models. In settings where data is strongly not missing at random, our methods achieve a 2-10% improvement in out-of-sample accuracy.

关键词： Missing data adaptive optimization

来源：评论

学校读者我要写书评

暂无评论

Physarum polycephalum-inspired adaptive optimization design of artificial microtubular networks

引用

Science China Chemistry 2025年第4期68卷 1497-1510页

作者： Si-Min Zeng Di-Wei Zheng Ting Pan Xing-Lan Ding Yun Zhang Chi-Hui Yang Si-Xue Cheng Xian-Zheng Zhang Key Laboratory of Biomedical Polymers of Ministry of Education&Department of Chemistry Wuhan UniversityWuhan 430072China

Natural biological microtubular networks have undergone adaptive evolutionary selection and may offer viable solutions to design challenges in artificial microtubular *** plasmodium of the slime mold Physarum polycephalum(***)extends continuously to form a protoplasmic microtubular network structure;directly connecting food *** simulations revealed that the formation of this adaptive *** microtubular network could be captured by a mathematical *** by the *** microtubular networks;we propose an adaptive optimization design method for artificial microtubular ***;we utilized hydrogels with biodegradable and tissue-adhesive properties to replicate the *** microtubular networks via *** Rhodamine B diffusion and glucosecatalyzed reaction experiments;we found that the *** microtubular networks exhibited significantly enhanced efficiency compared to vascular and artificial ***;we demonstrated the potential for uric acid(UA)degradation of the hydrogels with a real *** microtubular network loaded with urate oxidase(UOx)in a rodent model of *** this network achieved more than double the effect of the artificial *** underscores the potential of natural microtubular networks to replace artificial microtubular networks.

关键词： Physarum polycephalum adaptive optimization microtubular networks glucose-catalyzed reaction uric acid degradation

来源：评论

学校读者我要写书评

暂无评论

Unified convergence analysis for adaptive optimization with moving average estimator

引用

MACHINE LEARNING 2025年第4期114卷 1-51页

作者： Guo, Zhishuai Xu, Yi Yin, Wotao Jin, Rong Yang, Tianbao Texas A&M Univ College Stn TX 77840 USA Dalian Univ Technol Dalian 116024 Liaoning Peoples R China Alibaba Bellevue WA 98004 USA Twitter Seattle WA USA

Although adaptive optimization algorithms have been successful in many applications, there are still some mysteries in terms of convergence analysis that have not been unraveled. This paper provides a novel non-convex analysis of adaptive optimization to uncover some of these mysteries. Our contributions are three-fold. First, we show that an increasing or large enough momentum parameter for the first-order moment used in practice is sufficient to ensure the convergence of adaptive algorithms whose adaptive scaling factors of the step size are bounded. Second, our analysis gives insights for practical implementations, e.g., increasing the momentum parameter in a stage-wise manner in accordance with stagewise decreasing step size would help improve the convergence. Third, the modular nature of our analysis allows its extension to solving other optimization problems, e.g., compositional, min-max and bilevel problems. As an interesting yet non-trivial use case, we present algorithms for solving non-convex min-max optimization and bilevel optimization that do not require using large batches of data to estimate gradients or double loops as the literature do. Our empirical studies corroborate our theoretical results.

关键词： adaptive optimization Min-max problems Bilevel problems

来源：评论

学校读者我要写书评

暂无评论

Simulation-based adaptive optimization for passenger flow control measures at metro stations

引用

SIMULATION MODELLING PRACTICE AND THEORY 2025年 138卷

作者： Zhou, Yiqi Zhong, Maohua Li, Zhongwen Xu, Xuan Hua, Fucai Pan, Rongliang Tsinghua Univ Inst Publ Safety Res Dept Engn Phys Beijing 100084 Peoples R China Tsinghua Univ Beijing Urban Construct Design & Dev Grp Co Ltd Joint Res Ctr Urban Disaster Prevent & Safety Dept Engn Phys Beijing 100084 Peoples R China Shenzhen Metro Grp Co Ltd Shenzhen 518026 Peoples R China China Acad Safety Sci & Technol Beijing Key Lab Metro Fire & Passenger Transportat Beijing 100012 Peoples R China Beijing Urban Construct Design & Dev Grp Co Ltd Beijing 100037 Peoples R China

Effective passenger flow control measures are essential for the safe operation of metro stations. Existing in-station control measures include adjusting the operation mode of escalators and setting up temporary fences. However, in practice, metro operators often adopt fixed operation modes during fixed periods, indicating that the current passenger flow control measures at metro stations are overly rigidified. Therefore, developing an adaptive control strategy to constantly balance the wildly fluctuating passenger flow and optimize the operation performance is a key issue in current research. In this study, transportation efficiency and congestion risk are selected as evaluation objectives for passenger transportation risk, and passenger flow feature, station structure, and passenger flow control measures are considered key influential factors. Subsequently, an adaptive optimization method integrating simulation and data interpolation is proposed. The software Legion is used to conduct 150 orthogonal simulations, and prediction models for passenger transportation risk are obtained by performing data interpolation on the simulation results. Finally, taking a certain metro station as a case study, the optimal passenger flow control strategy under any passenger flow composition is obtained by scenario acquisition, risk identification, and adaptive decision-making. The results show that setting up temporary fences can reduce the passenger density near the fare gates, while adjusting the running direction of escalators can reduce overcrowding on the platform. Under varying passenger flow composition, the optimal strategy for the current scenario can be obtained, controlling passenger transportation risk within an acceptable range and providing assistance for metro operators in decision-making.

关键词： Metro station Pedestrian simulation adaptive optimization Passenger flow control measures Passenger transportation

来源：评论

学校读者我要写书评

暂无评论

Differentially private graph neural networks for graph classification and its adaptive optimization

引用

EXPERT SYSTEMS WITH APPLICATIONS 2025年 263卷

作者： Li, Yong Song, Xiao Gong, Kaiqi Liu, Songsong Li, Wenxin

Graph Neural Networks (GNNs), which outperform traditional deep learning algorithms in domains such as protein interaction prediction and molecular structure elucidation, have demonstrated superior performance in processing graph data. Despite the successful application of Differential Privacy Stochastic Gradient Descent (DPSGD) in deep learning, its adaptation to GNNs presents challenges due to the complexities inherent in GNN architectures and their distinctive message-passing mechanism. To solve the problem, this study delves into the challenges encountered when implementing DP-SGD in GNNs (DPGNNs), focusing on factors associated with the GNN model including the choice of learning rates, GNN frameworks, batch size, and gradient clipping thresholds, alongside parameters related to the Differential Privacy (DP) algorithm such as sensitivity, dataset size, privacy budget. The study reveals the impact of these factors on the performance of DPGNNs and provides recommendations for parameter values. Furthermore, we propose two adaptive optimization strategies to address the unique effects of learning rate and privacy budget on DPGNNs: adaptive learning rates and adaptive noise scales. These adaptive strategies automate parameter tuning, eliminating complex manual adjustments, significantly reducing training time and energy consumption, while achieving an optimal balance between privacy protection and model performance enhancement. Extensive experiments on graph classification tasks across multiple realworld datasets have validated the effectiveness of these strategies, significantly improving the classification accuracy and robustness of DPGNNs while ensuring data privacy.

关键词： Differential Privacy Graph Neural Network Graph Classification Stochastic Gradient Descent adaptive optimization Renyi differential privacy

来源：评论

学校读者我要写书评

暂无评论

adaptive optimization and dynamic pricing in decentralized energy markets using blockchain technology and consensus-based verification

引用

SUSTAINABLE ENERGY GRIDS & NETWORKS 2025年 42卷

作者： Naraindath, Nishkar R. Naidoo, Raj M. Bansal, Ramesh C. Univ Pretoria Dept Elect Elect & Comp Engn Gauteng South Africa Univ Sharjah Dept Elect Engn Sharjah U Arab Emirates

Peer-to-peer (P2P) markets are the key to unlocking the streamlined convergence of the prominent 5D elements in microgrids. However, current implementations focus on conventional methods that prioritize electricity cost reduction which often results in sub-optimal grid operation. This underscores the need for holistic and adaptive optimization in decentralized energy markets. This research introduces a novel consensus strategy built on principles from blockchains to serve as an overarching cross-verification tool that ensures integrity between off-chain and on-chain computations. The strategy leverages a dynamic stake function and reputation system to outperform traditional proof-of-stake. An adaptive optimization model along with a dynamic pricing model is then proposed and validated through multiple Python simulations. The work is proven to improve P2P interactions and grid efficiency. Furthermore, the overall system was implemented in a Solidity smart contract and deployed on an Ethereum test work to demonstrate the interoperability and functionality of the framework proposed. Suggestions for subsequent research are additionally included. In summary, this paper contributes to decentralized, equitable, efficient and self-sufficient microgrids.

关键词： adaptive optimization Blockchain technology Consensus-based verification Decentralized energy markets Dynamic pricing Microgrids Smart contracts

来源：评论

学校读者我要写书评

暂无评论

An adaptive optimization Method for Acoustic Temperature Measurement Topology Based on Multiple Sub-Objectives

引用

SENSORS 2025年第6期25卷 1878-1878页

作者： Zhu, Jialiang Zhou, Xinzhi Wang, Hailin Chen, Yixiao Xu, Tao He, Zhengxi Sichuan Univ Coll Elect & Informat Engn Chengdu 610065 Peoples R China Nucl Power Inst China Natl Key Lab Nucl Reactor Technol Chengdu 610213 Peoples R China

Recent years have seen a surge in study interest in acoustic temperature measurement because of its exceptional non-invasiveness, high precision, and fast response characteristics. Its main benefit is that it may rely on the temperature field reconstruction technique to obtain the entire temperature distribution information, circumventing the limitations of point-type thermometry. Studies have shown that the acoustic wave transducer topology is a key factor affecting the reconstruction effect. In engineering, a simple uniform placement or trial-and-error methods are often used to determine the transducer topology. However, these approaches lack adaptability in complex temperature fields, resulting in poor accuracy and stability. In this paper, based on the previous research on high-precision temperature field reconstruction algorithms, an adaptive optimization method of acoustic temperature measurement topology based on multiple sub-objectives is proposed. The method further improves the reconstruction of asymmetric complex temperature fields by constructing a new optimization variable and a new optimization objective. Comparison experiments with existing optimization methods demonstrate the effectiveness of the new variables and objectives. Additionally, the reconstruction performance of the proposed method is thoroughly evaluated. The results indicate that the method enables adaptive optimization of transducer topology. Moreover, the optimized results exhibit high accuracy and stability in reconstructing complex, asymmetric temperature fields.

关键词： acoustic tomography adaptive optimization asymmetric distribution temperature field reconstruction topology multi-objective

来源：评论

学校读者我要写书评

暂无评论

adaptive optimization of run-to-run controllers: The EWMA example

引用

IEEE TRANSACTIONS ON SEMICONDUCTOR MANUFACTURING 2000年第1期13卷 97-107页

作者： Patel, NS Jenkins, ST Texas Instruments Inc Dallas TX 75243 USA

This paper presents a recursive scheme for optimizing the gain of an exponentiallly weighted moving average (EWMA) controller under stability constraints. The objective is to minimize the asymptotic mean square error in the output with minimal a priori information. The algorithm hinges on a simple representation of the optimal EWMA gain. Both step and drift disturbances are considered. It is shown that the gain sequence generated by the algorithm always yields a stable system. Furthermore, this sequence is shown to converge to a suboptimal value. Extensions to the algorithm to the case where there is model uncertainty are also presented. The algorithm is verified via simulation. Data from a manufacturing implementation is presented.

关键词： adaptive optimization exponentially weighted moving average (EWMA) controller run-to-run control (R2R) stochastic approximation

来源：评论

学校读者我要写书评

暂无评论

adaptive optimization of face milling operations using neural networks

引用

JOURNAL OF MANUFACTURING SCIENCE AND ENGINEERING-TRANSACTIONS OF THE ASME 1998年第2期120卷 443-451页

作者： Ko, TJ Cho, DW Yeungnam Univ Dept Engn Mech Kyongsan 712749 Kyungbuk South Korea Pohang Univ Sci & Technol Dept Mech Engn Pohang 790784 Kyungbuk South Korea

In intelligent machine tools, a computer based control sq stem, which can adapt the machining parameters in an optimal fashion based on sensor measurements of the machining process, should be incorporated In this paper, the method for adaptive optimization of the cutting conditions in a face milling operation for maximizing the material removal rate is proposed. The optimization procedure described uses an exterior penalty function method in conjunction with a multilayered neural network. Two neural networks are introduced: one for estimating tool wear length, and the other for mapping input and output relations from the experimental data during cutting. The adaptive optimization of the cutting conditions is then implemented using the tool wear information and predicted process output. The results are demonstrated with respect to each level of machining such as rough, fine, and finish cutting.

关键词： penalty function adaptive optimization Mechanical Processes Cutting conditions Face milling control systems optimization Procedure Surgery precision machining material removal rate Neural network milling operation machining tool wear

来源：评论

学校读者我要写书评

暂无评论

adaptive optimization of a nitrifying sequencing batch reactor

引用

WATER RESEARCH 1999年第17期33卷 3569-3576页

作者： Katsogiannis, AN Kornaros, ME Lyberatos, GK Univ Patras Dept Comp Engn GR-26500 Patras Greece Inst Chem Engn & High Temp Chem Proc GR-26500 Patras Greece

Although substantial research effort has been expended towards the modeling of nitrification process, optimization and control of nitrifying systems, especially using a sequencing batch reactor (SBR) configuration, is still lacking appropriate consideration. The present work addresses the problem of minimizing the cost of nitrification, using an on-line adaptive optimization algorithm. The algorithm is based on the on-line identification of the batch cycle duration and is used for the on-line determination of the fraction of switching period to be implemented. The performance of the algorithm was tested against a simplified nitrification model adopted from the activated sludge model No. 2 (ASM2). The convergence of the algorithm to the optimum fraction of switching period, as well as its ability to satisfactorily handle disturbances in the influent ammonium-nitrogen concentration, is demonstrated. (C) 1999 Elsevier Science Ltd. All rights reserved.

关键词： adaptive control adaptive optimization nitrification SBR sequencing batch reactor

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：