检索结果-内蒙古大学图书馆

Enhancing multiview synergy: Robust learning by exploiting the wave loss function with consensus and complementarity principles

引用

NEURAL NETWORKS 2025年 188卷 107433页

作者： Quadir, A. Akhtar, Mushir Tanveer, M. Indian Inst Technol Indore Dept Math Indore 453552 India

Multiview learning (MvL) is an advancing domain in machine learning, leveraging multiple data perspectives to enhance model performance through view-consistency and view-discrepancy. Despite numerous successful multiview-based support vector machine (SVM) models, existing frameworks predominantly focus on the consensus principle, often overlooking the complementarity principle. Furthermore, they exhibit limited robustness against noisy, error-prone, and view-inconsistent samples, prevalent in multiview datasets. To tackle the aforementioned limitations, this paper introduces Wave-MvSVM, a novel multiview support vector machine framework leveraging the wave loss (W-loss) function, specifically designed to harness both consensus and complementarity principles. Unlike traditional approaches that often overlook the complementary information among different views, the proposed Wave-MvSVM ensures a more comprehensive and resilient learning process by integrating both principles effectively. The W-loss function, characterized by its smoothness, asymmetry, and bounded nature, is particularly effective in mitigating the adverse effects of noisy and outlier data, thereby enhancing model stability. Theoretically, the W-loss function also exhibits a crucial classification-calibrated property, further boosting its effectiveness. The proposed Wave-MvSVM employs a between-view co-regularization term to enforce view consistency and utilizes an adaptive combination weight strategy to maximize the discriminative power of each view, thus fully exploiting both consensus and complementarity principles. The optimization problem is efficiently solved using a combination of gradient descent (GD) and the alternating direction method of multipliers (admm), ensuring reliable convergence to optimal solutions. The generalization abilities of the proposed Wave-MvSVM model is theoretically supported through analyses based on Rademacher complexity. Extensive empirical evaluations across diverse da

关键词： Multiview learning (MvL) Wave loss function Classification-calibration Consensus and complementary information admm algorithm Rademacher complexity

来源：评论

学校读者我要写书评

暂无评论

Single-Shot Phase Retrieval Via Gradient-Sparse Non-Convex Regularization Integrating Physical Constraints

引用

JOURNAL OF SCIENTIFIC COMPUTING 2025年第3期102卷 1-41页

作者： Chen, Xuesong Li, Fang East China Normal Univ Sch Math Sci Shanghai 200241 Peoples R China East China Normal Univ Sch Math Sci Key Lab MEA Minist Educ Shanghai 200241 Peoples R China East China Normal Univ Shanghai Key Lab PMMP Shanghai 200241 Peoples R China

Measurements of light typically capture amplitude information, often overlooking crucial phase details. This oversight underscores the importance of phase retrieval (PR), essential in biomedical imaging, X-ray crystallography, and microscopy, for reconstructing complex signals from intensity-only measurements. Traditional methods frequently fall short, especially in noisy conditions or when restricted to single-shot measurements. To address these challenges, we introduce a novel model that combines non-convex regularization with physical constraints. The model adopts the smoothly clipped absolute deviation (SCAD) function as a sparsity regularization term for gradients, incorporating fundamental constraints on support and absorption to establish an inverse model. Using the alternating direction method of multipliers (admm), we break down the problem into manageable sub-problems, implementing SCAD shrinkage in the complex domain and applying Wirtinger gradient projection methods. A thorough convergence analysis validates the stability and robustness of the algorithm. Extensive simulations confirm significant improvements in reconstruction quality compared to existing methods, with evaluations demonstrating superior performance across various noise levels and parameter settings.

关键词： Phase retrieval SCAD function admm algorithm Wirtinger gradient Non-convex optimization

来源：评论

学校读者我要写书评

暂无评论

Distributed estimation for linear regression with covariates missing at random

引用

COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION 2025年第2期54卷 583-601页

作者： Pan, Yingli Wang, Haoyu Xu, Kaidong Huang, He Hubei Univ Fac Math & Stat Wuhan Peoples R China Wuzhou Univ Fac Management Wuzhou Peoples R China

In this article, we propose a distributed optimization approach to analyzing distributed-stored data based on linear regression with covariates missing at random. To solve distributed statistical learning problems, we construct a communication-efficient surrogate weighted loss function for the global weighted loss function based on inverse probability-weighted and communication-efficient surrogate loss method. Combining the advantages of the surrogate weighted loss function with alternating direction method of multipliers algorithm, a distributed algorithm is developed for the calculation of our proposed estimator. Under some mild assumptions, we establish the asymptotic properties of the proposed estimator. Simulation studies are performed to access the finite-sample performance of our method. We further illustrate the advantages of our method by a real-world dataset from Beijing Municipal Environmental Monitoring Center.

关键词： admm algorithm CSL Distributed optimization IPW MAR

来源：评论

学校读者我要写书评

暂无评论

Tensor Decomposition-assisted Multiview Subgroup Analysis

引用

Acta Mathematica Sinica,English Series 2025年第2期41卷 588-618页

作者： Xun Zhao Ling Zhou Weijia Zhang Huazhen Lin Center of Statistical Research and School of Statistics New Cornerstone Science LaboratorySouthwestern University of Finance and EconomicsChengdu 610064P.R.China Department of Financial and Actuarial Mathematics Xi'an Jiaotong-Liverpool UniversitySuzhou 215123P.R.China

To learn the subgroup structure generated by multidimensional interaction, we propose a novel multiview subgroup integration technique based on tensor decomposition. Compared to the traditional subgroup analysis that can only handle single-view heterogeneity, our proposed method achieves a greater level of homogeneity within the subgroups, leading to enhanced interpretability and predictive power. For computational readiness of the proposed method, we build an algorithm that incorporates pairwise shrinkage-encouraging penalties and admm techniques. Theoretically, we establish the asymptotic consistency and normality of the proposed estimators. Extensive simulation studies and real data analysis demonstrate that our proposal outperforms other methods in terms of prediction accuracy and grouping consistency. In addition, the analysis based on the proposed method indicates that intergenerational care significantly increases the risk of chronic diseases associated with diet and fatigue in all provinces while only reducing the risk of emotion-related chronic diseases in the eastern coastal and central regions of China.

关键词： Multiview subgroup analysis tensor decomposition data integration admm algorithm

来源：评论

学校读者我要写书评

暂无评论

A flexible and efficient algorithm for high dimensional support vector regression

引用

NEUROCOMPUTING 2025年 611卷

作者： Yang, Menglei Liang, Hao Wu, Xiaofei Zhang, Zhimin Chongqing Univ Coll Math & Stat Chongqing 401331 Peoples R China Shandong Univ Cheeloo Coll Med Sch Publ Hlth Dept Microbiol Jinan 250012 Peoples R China

In high dimensional statistical learning, variable selection and handling highly correlated phenomena are two crucial topics. Elastic-net regularization can automatically perform variable selection and tends to either simultaneously select or remove highly correlated variables. Consequently, it has been widely applied in machine learning. In this paper, we incorporate elastic-net regularization into the support vector regression model, introducing the Elastic-net Support Vector Regression (En-SVR) model. Due to the inclusion of elastic- net regularization, the En-SVR model possesses the capability of variable selection, addressing high dimensional and highly correlated statistical learning problems. However, the optimization problem for the En-SVR model is rather complex, and common methods for solving the En-SVR model are challenging. Nevertheless, we observe that the optimization problem for the En-SVR model can be reformulated as a convex optimization problem where the objective function is separable into multiple blocks and connected by an inequality constraint. Therefore, we employ a novel and efficient Alternating Direction Method of Multipliers (admm) algorithm to solve the En-SVR model, and provide a complexity analysis as well as convergence analysis for the algorithm. Furthermore, extensive numerical experiments validate the outstanding performance of the En-SVR model in high dimensional statistical learning and the efficiency of this novel admm algorithm.

关键词： Support vector regression Elastic-net regularization admm algorithm Variable selection High dimensional statistical learning

来源：评论

学校读者我要写书评

暂无评论

Multicore and Manycore Implementations of admm-based Decoders for LDPC Decoding

引用

JOURNAL OF SIGNAL PROCESSING SYSTEMS FOR SIGNAL IMAGE AND VIDEO TECHNOLOGY 2018年第11期90卷 1551-1567页

作者： Debbabi, Imen Le Gal, Bertrand Khouja, Nadia Tlili, Fethi Jego, Christophe Carthage Univ High Sch Commun GRESCOM Lab Ariana Tunisia Bordeaux Univ Bordeaux INP CNRS Lab IMSUMR 5218 Bordeaux France

The alternate direction method of multipliers (admm) algorithm has recently been proposed for LDPC decoding based on linear programming (LP) techniques. Even though it improves the error rate performance compared with usual message passing (MP) techniques, it shows a higher computation complexity. However, a significant step towards LP LDPC decoding scalability and optimization is made possible since the admm algorithm acts as an MP decoding one. In this paper, an overview of the admm approach and its error correction performances is provided. Then, its computation and memory complexities are evaluated. Finally, optimized software implementations of the decoder to take advantage of multi/many-core device features are described. Optimization choices are discussed and justified according to execution profiling figures and the algorithm's parallelism levels. Experimentation results show that this LP based decoding technique can reach WiMAX and WRAN standards real time throughput requirements on mid-range devices.

关键词： Many-core Multi-core LDPC decoding Linear programming admm algorithm Software optimization SIMD SIMT GPU Throughput

来源：评论

学校读者我要写书评

暂无评论

Asymmetric kernel-based robust classification by admm

引用

KNOWLEDGE AND INFORMATION SYSTEMS 2023年第1期65卷 89-110页

作者： Ding, Guangsheng Yang, Liming China Agr Univ Coll Sci Beijing 100083 Peoples R China

Correntropy is a locally second-order statistical measure in kernel space. The different kernel functions induce different correntropy with different properties. In this work, we propose an asymmetric mixture kernel and the corresponding correntropy. Then, we propose a new loss function (called RCH-loss) that is induced by the correntropy with the reproducing asymmetric kernel. Some important properties of the proposed kernel and RCH-loss are demonstrated such as non-convexity, boundedness, asymmetry and asymptotic approximation. Moreover, the proposed RCH-loss satisfies Bayes optimal decision rule. With the RCH-loss function, a new robust classification framework is built to handle robust classification. Theoretically, we prove the generalization bound of the proposed model based on the Rademacher complexity. Following that, DC (difference of convex functions) programming algorithm (DCA) is developed to solve the problem iteratively, where admm (alternating direction method of multipliers) is used to solve the subproblem at each iteration. Moreover, we analyze the computation complexity of the algorithm and the sensitivity of parameters. Numerical experimentations are carried out on various datasets including benchmark data sets and artificial data sets with different noise levels. The experimental results display the feasibility and effectiveness of the proposed methods.

关键词： Robustness Asymmetric mixture kernel Correntropy admm algorithm DC programming algorithm Generalization bound

来源：评论

学校读者我要写书评

暂无评论

Joint Sparsity and Order Optimization Based on admm With Non-Uniform Group Hard Thresholding

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS 2018年第5期65卷 1602-1613页

作者： Matsuoka, Ryo Kyochi, Seisuke Ono, Shunsuke Okuda, Masahiro Kagawa Univ Fac Engn Takamatsu Kagawa 7610396 Japan Univ Kitakyushu Fac Environm Engn Kitakyushu Fukuoka 8080135 Japan Tokyo Inst Technol Lab Future Interdisciplinary Res Sci & Technol Yokohama Kanagawa 2268503 Japan

This paper proposes a new optimization framework for the joint optimization of sparsity and filter order (JOSFO) for FIR filter design. Since the cost function for JOSFO involves l(0) and non-uniform overlapped group l(0) norms, which are not convex, a global optimal solution is difficult to obtain. To find an approximate solution of the non-convex problem, existing approaches repeat the following steps: 1) approximate the cost function;2) find candidates of zero coefficients by minimizing the cost function;and 3) set them to zero. On the other hand, this paper directly solves the optimization problem, without any approximation to the cost function, by using the alternating direction method of multipliers with the pseudo-proximity operators of l(0) and non-uniform non-overlapped group l(0) norms. Experimental results show that resulting filters designed by the proposed method have sparser coefficients and lower orders, while satisfying filter specifications, such as an error from a desired frequency response.

关键词： FIR filter design sparsity l(0) pseudo-norm admm algorithm

来源：评论

学校读者我要写书评

暂无评论

admm for monotone operators: convergence analysis and rates

引用

ADVANCES IN COMPUTATIONAL MATHEMATICS 2019年第1期45卷 327-359页

作者： Bot, Radu Ioan Csetnek, Ernoe Robert Univ Vienna Fac Math Oskar Morgenstern Pl 1 A-1090 Vienna Austria

We propose in this paper a unifying scheme for several algorithms from the literature dedicated to the solving of monotone inclusion problems involving compositions with linear continuous operators in infinite dimensional Hilbert spaces. We show that a number of primal-dual algorithms for monotone inclusions and also the classical admm numerical scheme for convex optimization problems, along with some of its variants, can be embedded in this unifying scheme. While in the first part of the paper, convergence results for the iterates are reported, the second part is devoted to the derivation of convergence rates obtained by combining variable metric techniques with strategies based on suitable choice of dynamical step sizes. The numerical performances, which can be obtained for different dynamical step size strategies, are compared in the context of solving an image denoising problem.

关键词： Monotone operators Primal-dual algorithm admm algorithm Subdifferential Convex optimization Fenchel duality

来源：评论

学校读者我要写书评

暂无评论

Application of Least Square Denoising to improve admm based Hyperspectral Image Classification

Application of Least Square Denoising to improve ADMM based ...

引用

6th International Conference on Advances in Computing and Communications (ICACC)

作者： Srivatsa, S. Ajay, Aleena Chandni, C. K. Sowmya, V. Soman, K. P. Amrita Univ Amrita Vishwa Vidyapeetham Amrita Sch Engn Ctr Computat Engn & Networking CEN Coimbatore Tamil Nadu India

Hyperspectral images contain a huge amount of spatial and spectral information so that, almost any type of Earth feature can be discriminated from any other feature. But, for this classification to be possible, it is to be ensured that there is as less noise as possible in the captured data. Unfortunately, noise is unavoidable in nature and most hyperspectral images need denoising before they can be processed for classification work. In this paper, we are presenting a new approach for denoising hyperspectral images based on Least Square Regularization. Then, the hyperspectral data is classified using Basis Pursuit classifier, a constrained L1 minimization problem. To improve the time requirement for classification, Alternating Direction Method of Multipliers (admm) solver is used instead of CVX (convex optimization) solver. The method proposed is compared with other existing denoising methods such as Legendre-Fenchel (LF), Wavelet thresholding and Total Variation (TV). It is observed that the proposed Least Square (LS) denoising method improves classification accuracy much better than other existing denoising techniques. Even with fewer training sets, the proposed denoising technique yields better classification accuracy, thus proving least square denoising to be a powerful denoising technique. (C) 2016 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://***/licenses/by-nc-nd/4.0/).

关键词： Denoising hyperspectral image classification basis pursuit admm algorithm Least Square Legendre-Fenchel transformation

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：