检索结果-内蒙古大学图书馆

High-Dimensional Nonconvex Stochastic optimization by Doubly Stochastic Successive Convex Approximation

IEEE TRANSACTIONS ON SIGNAL PROCESSING 2020年 68卷 6287-6302页

作者： Mokhtari, Aryan Koppel, Alec Univ Texas Austin Dept Elect & Comp Engn Austin TX 78712 USA US Army Computat & Informat Sci Directorate Res Lab Adelphi MD 20783 USA

In this paper, we consider supervised learning problems over training sets in which the number of training examples and the dimension of feature vectors are both large. We focus on the case where the loss function defining the quality of the parameter we wish to estimate may be non-convex, but also has a convex regularization. We propose a Doubly Stochastic Successive Convex approximation scheme (DSSC) able to handle non-convex regularized expected risk minimization. The method operates by decomposing the decision variable into blocks and operating on random subsets of blocks at each step (fusing the merits of stochastic approximation with block coordinate methods), and then implements successive convex approximation. In contrast to many stochastic convex methods whose almost sure behavior is not guaranteed in non-convex settings, DSSC attains almost sure convergence to a stationary solution of the problem. Moreover, we show that the proposed DSSC algorithm achieves stationarity at a rate of O((log t)/t(1/4)). Numerical experiments on a non-convex variant of a lasso regression problem show that DSSC performs favorably in this setting. We then apply this method to the task of dictionary learning from high-dimensional visual data collected from a ground robot, and observe reliable convergence behavior for a difficult non-convex stochastic program.

关键词： optimization algorithms stochastic optimization successive convex approximation

来源：评论

学校读者我要写书评

暂无评论

Sparse data-driven knowledge discovery for interpretable prediction of permeability in tight sandstones

引用

Engineering Geology 2025年 353卷

作者： Xu, Lulu Du, Zhengyang Cai, Meifeng Yin, Shangxian Dong, Shuning Thanh, Hung Vo Carroll, Kenneth C. Soltanian, Mohamad Reza Dai, Zhenxue College of Construction Engineering Jilin University State Key Laboratory of Deep Earth Exploration and Imaging Changchun China Institute of Intelligent Simulation and Early Warning for Subsurface Environment Jilin University Changchun130026 China School of Civil and Resource Engineering University of Science and Technology Beijing Beijing100083 China College of Safety Engineering North China Institute of Science and Technology Langfang065201 China Xi'an Research Institute Co. Ltd. China Coal Technology and Engineering Group Corp Xi'an710077 China Waseda Research Institute for Science and Engineering Waseda University 3-4-1 Okubo Tokyo Shinjuku-ku169-8555 Viet Nam Department of Plant and Environmental Sciences New Mexico State University Las CrucesNM88003 United States Department of Geosciences University of Cincinnati CincinnatiOH45220 United States Department of Environmental Engineering University of Cincinnati CincinnatiOH45220 United States School of Environmental and Municipal Engineering Qingdao University of Technology Qingdao266033 China

Permeability (k) is crucial for subsurface fluid flow, but predicting k-values in tight sandstones remains challenging due to their complex pore structure and heterogeneity. Although machine learning (ML) has shown promise, it faces significant challenges, including limited high-quality data, high computational costs, and unclear prediction mechanisms. This study proposes a sparse data-driven knowledge discovery framework aimed at enhancing the accuracy and interpretability of k-value predictions in tight sandstone formations. We integrate ML models with data augmentation (ML-DA), using Extreme Gradient Boosting (XGBoost-DA) and Least Squares Support Vector Regression (LSSVR-DA), optimized through genetic algorithms (GA), particle swarm optimization (PSO), and Bayesian optimization (BO). SHapley Additive Explanations (SHAP) are employed to elucidate the interactions between key factors influencing predictions. Monte Carlo simulations demonstrate the robust performance of our ML-DA models, even under data constraints. SHAP analysis identifies key predictors, including porosity, displacement pressure, median pore throat radius, median pressure, and carbonate content. Partial dependence plots (PDPs) reveal a significant interaction between porosity and carbonate content, as well as a decrease in model stability at low carbonate content. This study presents an interpretable ML framework with data augmentation, enabling improved predictions from sparse data while exploring the interactions between key factors. The framework can be adapted to other domains facing similar challenges, enhancing the accuracy and transparency of model predictions. © 2025 Elsevier B.V.

关键词： Data augmentation Explainable AI Hybrid intelligence optimization algorithms Permeability prediction Tight sandstones

来源：评论

学校读者我要写书评

暂无评论

Optimizing the Cost and Reliability of Shared Anchors in an Array of Floating Offshore Wind Turbines

引用

ASCE-ASME JOURNAL OF RISK AND UNCERTAINTY IN ENGINEERING SYSTEMS PART B-MECHANICAL ENGINEERING 2021年第4期7卷 040905页

作者： Devin, Michael C. DuPont, Bryony L. Hallowell, Spencer T. Arwade, Sanjay R. Oregon State Univ Sch Mech Ind & Mfg Engn Corvallis OR 97331 USA Baker Design Consultants Freeport ME 04032 USA Univ Massachusetts Amherst MA 01003 USA

Commercial floating offshore wind projects are expected to emerge in the U.S. by the end of this decade. Currently, however, high costs for the technology limit its commercial viability, and a lack of data regarding system reliability heightens project risk. This work presents an optimization algorithm to examine the tradeoffs between cost and reliability for a floating offshore wind array that uses shared anchoring. Combining a multivariable genetic algorithm with elements of Bayesian optimization, the optimization algorithm selectively increases anchor strengths to minimize the added costs of failure for a large floating wind farm in the Gulf of Maine under survival load conditions. The algorithm uses an evaluation function that computes the probability of mooring system failure, then calculates the expected maintenance costs of a failure via a Monte Carlo method. A cost sensitivity analysis is also performed to compare results for a range of maintenance cost profiles. The results indicate that virtually all of the farm's anchors are strengthened in the minimum cost solution. Anchor strength is increased between 5 and 35% depending on farm location, with anchor strength nearest the export cable being increased the most. The optimal solutions maintain a failure probability of 1.25%, demonstrating the tradeoff point between cost and reliability. System reliability was found to be particularly sensitive to changes in turbine costs and downtime, suggesting further research into floating offshore wind turbine failure modes in extreme loading conditions could be particularly impactful in reducing project uncertainty.

关键词： Floating wind turbines Ocean engineering optimization Turbines Stress Genetic algorithms Uncertainty Reliability Wind Wind farms Failure optimization algorithms Mooring Maintenance algorithms Cables

来源：评论

学校读者我要写书评

暂无评论

Bridging Evolutionary algorithms and Reinforcement Learning: A Comprehensive Survey on Hybrid algorithms

引用

IEEE Transactions on Evolutionary Computation 2024年 1-1页

作者： Li, Pengyi Hao, Jianye Tang, Hongyao Fu, Xian Zhen, Yan Tang, Ke College of Intelligence and Computing Tianjin University Tianjin China Quebec Canada Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen China

Evolutionary Reinforcement Learning (ERL), which integrates Evolutionary algorithms (EAs) and Reinforcement Learning (RL) for optimization, has demonstrated remarkable performance advancements. By fusing both approaches, ERL has emerged as a promising research direction. This survey offers a comprehensive overview of the diverse research branches in ERL. Specifically, we systematically summarize recent advancements in related algorithms and identify three primary research directions: EA-assisted optimization of RL, RL-assisted optimization of EA, and synergistic optimization of EA and RL. Following that, we conduct an in-depth analysis of each research direction, organizing multiple research branches. We elucidate the problems that each branch aims to tackle and how the integration of EAs and RL addresses these challenges. In conclusion, we discuss potential challenges and prospective future research directions across various research directions. To facilitate researchers in delving into ERL, we organize the algorithms and codes involved on https://***/yeshenpy/Awesome-Evolutionary-Reinforcement-Learning IEEE

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Research on Computer-aided Innovation and Entrepreneurship Education Model and Algorithm

引用

Applied Mathematics and Nonlinear Sciences 2024年第1期9卷

作者： Wang, Jingru Xinyu University Jiangxi Xinyu338000 China

With the rapid development of science and technology, computer-aided technology has been widely used in the field of education, but in the field of innovation and entrepreneurship education (Hereinafter referred to as IEE), there are still problems such as uneven distribution of resources and the disconnection between teaching content and actual *** paper puts forward a new model framework of computer-aided IEE, which is student-centered and *** clear teaching objectives, rich course content, diverse teaching methods, and a scientific evaluation system, students' innovative consciousness and entrepreneurial ability are comprehensively *** course covers innovative thinking, business models, market research, and other theoretical knowledge, and introduces virtual reality (VR) and augmented reality (AR) technologies to simulate the real entrepreneurial process and provide a practical *** the aspect of algorithm application, the research combines the content recommendation algorithm and collaborative filtering recommendation algorithm, constructs accurate user portraits by analyzing students' learning behavior data, and matches them according to the characteristics of learning resources to recommend personalized learning resources for *** the same time, the learning path optimization algorithm is developed, and the personalized and efficient learning path is planned by using the shortest path algorithm in graph theory to ensure that students can master the required knowledge in the shortest *** experimental part designed a one-semester teaching experiment, and compared the differences between the experimental group and the control group in academic performance, learning participation, and *** results show that the computer-aided IEE model significantly improves students' academic performance, learning participation, and satisfaction, which verifies the effectiveness of the *** research on the

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Quantum Marine Predator Algorithm: A Quantum Leap in Photovoltaic Efficiency Under Dynamic Conditions

引用

INFORMATION 2024年第11期15卷 692页

作者： Fergani, Okba Himeur, Yassine Mechgoug, Raihane Atalla, Shadi Mansoor, Wathiq Tkouti, Nacira Univ Biskra Lab Identificat Commande Control & Commun LI3CUB Biskra 07000 Algeria Univ Dubai Coll Engn & Informat Technol POB 14143 Dubai U Arab Emirates Univ Biskra Dept Elect Engn Biskra 07000 Algeria

The Quantum Marine Predator Algorithm (QMPA) presents a groundbreaking solution to the inherent limitations of conventional Maximum Power Point Tracking (MPPT) techniques in photovoltaic systems. These limitations, such as sluggish response times and inadequate adaptability to environmental fluctuations, are particularly pronounced in regions with challenging weather patterns like Sunderland. QMPA emerges as a formidable contender by seamlessly integrating the sophisticated hunting tactics of marine predators with the principles of quantum mechanics. This amalgamation not only enhances operational efficiency but also addresses the need for real-time adaptability. One of the most striking advantages of QMPA is its remarkable improvement in response time and adaptability. Compared to traditional MPPT methods, which often struggle to keep pace with rapidly changing environmental factors, QMPA demonstrates a significant reduction in response time, resulting in up to a 30% increase in efficiency under fluctuating irradiance conditions for a resistive load of 100 Omega. These findings are derived from extensive experimentation using NASA's worldwide power prediction data. Through a detailed comparative analysis with existing MPPT methodologies, QMPA consistently outperforms its counterparts, exhibiting superior operational efficiency and stability across varying environmental scenarios. By substantiating its claims with concrete data and measurable improvements, this research transcends generic assertions and establishes QMPA as a tangible advancement in MPPT technology.

关键词： Quantum Marine Predator Algorithm (QMPA) Maximum Power Point Tracking (MPPT) photovoltaic systems quantum mechanics optimization algorithms renewable energy efficiency

来源：评论

学校读者我要写书评

暂无评论

A Randomized Block-Coordinate Adam online learning optimization algorithm

引用

NEURAL COMPUTING & APPLICATIONS 2020年第16期32卷 12671-12684页

作者： Zhou, Yangfan Zhang, Mingchuan Zhu, Junlong Zheng, Ruijuan Wu, Qingtao Henan Univ Sci & Technol Sch Informat Engn Luoyang 471023 Peoples R China

In recent years, stochastic gradient descent (SGD) becomes one of the most important optimization algorithms in many fields, such as deep learning and reinforcement learning. However, the computation of full gradient in SGD is prohibitive when dealing with high-dimensional vectors. For this reason, we propose a randomized block-coordinate Adam (RBC-Adam) online learning optimization algorithm. At each round, RBC-Adam randomly chooses a variable from a subset of parameters to compute the gradient and updates the parameters along the negative gradient direction. Moreover, this paper analyzes the convergence of RBC-Adam and obtains the regret bound,O(T) whereTis a time horizon. The theoretical results are verified by simulated experiments on four public datasets. Moreover, the simulated experiment results show that the computational cost of RBC-Adam is lower than the variants of Adam.

关键词： Adam Convergence Online learning optimization algorithms Randomized block-coordinate Stochastic gradient descent

来源：评论

学校读者我要写书评

暂无评论

A Multiform Many-objective Evolutionary Algorithm for Multi-label Feature Selection in Classification

引用

IEEE Transactions on Evolutionary Computation 2024年

作者： Hancer, Emrah Xue, Bing Zhang, Mengjie Victoria University of Wellington Center for Data Science and Artificial Intelligence School of Engineering and Computer Science Wellington6140 New Zealand Burdur Mehmet Akif Ersoy University Department of Software Engineering Burdur15039 Turkey

Multi-label classification (MLC) deals with instances associated with multiple labels simultaneously and often includes high dimensional data with noisy, irrelevant, and redundant features. Feature selection for MLC is crucial for achieving successful classification performance. However, no unified metric exists for evaluating learning performance in MLC tasks. Instead, multiple metrics exist, each assessing a different aspect of the classification process, and these metrics are often inconsistent with one another. Consequently, multi-label feature selection (MLFS) becomes a many-objective optimization problem (MaOP) when optimizing three or more classification metrics and the number of selected features simultaneously. Evolutionary computation (EC) techniques have shown great promise in addressing many-objective tasks due to their ability to effectively explore large and complex search spaces. EC techniques can handle multiple objectives concurrently, making them well-suited for the challenges posed by MLFS. Despite this potential, research on EC-based MLFS methods remains limited, with few studies treating it as a MaOP. To address this gap, this paper proposes a new many-objective evolutionary algorithm within a multiform framework. The proposed algorithm leverages distinct sub-populations to address specific MLFS tasks, incorporates a strategy to exchange information between MLFS tasks, employs a local density-based selection mechanism to maintain diverse high-quality solutions, and uses an adaptive parameter scheme inspired by stochastic gradient descent to guide the search towards new regions in the objective space. Experimental results demonstrate the superiority of the proposed algorithm across a diverse range of high-dimensional datasets, outperforming both recently introduced many-objective and conventional MLFS algorithms. © 1997-2012 IEEE.

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Incremental specialized and specialized-generalized matrix factorization models based on adaptive learning rate optimizers

引用

NEUROCOMPUTING 2023年第1期552卷

作者： Viniski, Antonio David Barddal, Jean Paul Britto Jr, Alceu de Souza de Campos, Humberto Vinicius Aparecido Pontificia Univ Catolica Parana PUCPR Grad Program Informat PPGIa Rua Imaculada Conceicao1155 BR-80215901 Curitiba Brazil Himarket Ave Visconde Guarapuava2764 Curitiba Brazil

Recommender systems suggest items that are likely to be preferred by a particular user based on historical behavior, actions, and feedback. In real-world applications, data on users and items are continuously generated at a fast pace, such as in e-commerce, social media, digital marketing, and content consumption applications. Since interactions occur over time, these scenarios can be formulated as a data stream where users' interests are potentially dynamic, i.e., they change over time. Given that changes are expected to occur, one of the current research challenges in streaming recommender systems is that models must adapt their parameters when changes occur to maintain performance. As such changes do not occur for all users and items in the stream at the same time, we consider adapting learning schemes to account for user or item identifiers and model individual parameters. Therefore, we used specialized parameters to adjust the step size for each dataset user or item. More specifically, this study proposes four specialized and specialized-generalized variants of four well-known adaptive learning rate optimizers and shows how they are combined with incremental matrix factorization methods. We tested our proposed optimization strategies on different datasets and showed that one of the proposed specialized variants, that is, InAMSGradUser, improves the RECALL and NDCG rates by up to 11.1 and 7.5 percentage points, respectively, compared to the traditional stochastic gradient descent (SGD) optimizer. & COPY;2023 Elsevier B.V. All rights reserved.

关键词： Matrix factorization optimization algorithms Collaborative filtering

来源：评论

学校读者我要写书评

暂无评论

An innovative grey wolf optimizer with Nelder-mead search method based MPPT technique for fast convergence under partial shading conditions

引用

SUSTAINABLE ENERGY TECHNOLOGIES AND ASSESSMENTS 2023年 59卷

作者： Swetha, K. T. Reddy, Venugopal Robinson, Abin Natl Inst Technol Goa Dept Elect & Elect Engn Ponda 403401 Goa India

Partial shading condition (PSC) is an unavoidable problem faced by photovoltaic arrays, where multiple peaks appear on the power-voltage (P-V) curve. The conventional techniques are unable to track the global power under PSCs, and thus, various optimization techniques aim to reduce the impact of PSCs. This paper proposes a hybrid Grey wolf optimization with Nelder-mead (GWO-NM) algorithm for tracking the maximum power point. The GWO algorithm assures achieving the GMPP through search space exploration. In contrast, NM is a direct search technique that aids all particles in rapidly converging on the GMPP. Consequently, the proposed hybrid method has the ability to track actual GMPP with high efficiency, providing an enhanced convergence rate and fast response time, and is highly capable of operating under varying irradiance. The proposed algorithm avoids unnecessary exploration of particles and also effectively reduces steady-state oscillations. The proposed algorithm is experimentally validated using a DSP controller, showing an average tracking time of 0.54 s with a steady-state efficiency of 99.91% under different shading conditions. Finally, the effectiveness of the proposed GWO-NM algorithm is compared to recent meta-heuristic algorithms in terms of tracking power, speed, and efficiency.

关键词： optimization algorithms Photovoltaic systems Partial shading conditions Grey wolf optimization Maximum power point tracking Nelder-mead method

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：