检索结果-内蒙古大学图书馆

Sparse and robust support vector machine with capped squared loss for large-scale pattern classification

PATTERN RECOGNITION 2024年 153卷

作者： Wang, Huajun Zhang, Hongwei Li, Wenqian Changsha Univ Sci & Technol Dept Math & Stat Changsha Peoples R China Hunan Normal Univ Coll Life Sci Changsha Peoples R China

Support vector machine (SVM), being considered one of the most efficient tools for classification, has received widespread attention in various fields. However, its performance is hindered when dealing with large-scale pattern classification tasks due to high memory requirements and running very slow. To address this challenge, we construct a novel sparse and robust SVM based on our newly proposed capped squared loss (named as L-csl-SVM). To solve L-csl-SVM, we first focus on establishing optimality theory of L-csl-SVM via our defined proximal stationary point, which is convenient for us to efficiently characterize the L-csl support vectors of L-csl-SVM. We subsequently demonstrate that the L-csl support vectors comprise merely a minor fraction of entire training data. This observation leads us to introduce the concept of the working set. Furthermore, we design a novel subspace fast algorithm with working set (named as L-csl-ADMM) for solving L-csl-SVM, which is proven that L-csl-ADMM has both global convergence and relatively low computational complexity. Finally, numerical experiments show that L-csl-ADMM has excellent performances in terms of getting the best classification accuracy, using the shortest time and presenting the smallest numbers of support vectors when solving large-scale pattern classification problems.

关键词： Capped squared loss Fast algorithm Support vectors low computational complexity Working set

来源：评论

学校读者我要写书评

暂无评论

A novel neural dynamical approach to convex quadratic program and its efficient applications

引用

NEURAL NETWORKS 2009年第10期22卷 1463-1470页

作者： Xia, Youshen Sun, Changyin Fuzhou Univ Coll Math & Comp Sci Fuzhou Peoples R China Southeast Univ Dept Automat Nanjing Peoples R China

This paper proposes a novel neural dynamical approach to a class of convex quadratic programming problems where the number of variables is larger than the number of equality constraints. The proposed continuous-time and proposed discrete-time neural dynamical approach are guaranteed to be globally convergent to an optimal solution. Moreover, the number of its neurons is equal to the number of equality constraints. In contrast, the number of neurons in existing neural dynamical methods is at least the number of the variables. Therefore, the proposed neural dynamical approach has a low computational complexity. Compared with conventional numerical optimization methods, the proposed discrete-time neural dynamical approach reduces multiplication operation per iteration and has a large computational step length. computational examples and two efficient applications to signal processing and robot control further confirm the good performance of the proposed approach. (C) 2009 Elsevier Ltd. All rights reserved.

关键词： Neural dynamical approach Convex quadratic program low computational complexity Large computational step length

来源：评论

学校读者我要写书评

暂无评论

Rich QR code with three-layer information using visual secret sharing scheme

引用

MULTIMEDIA TOOLS AND APPLICATIONS 2019年第14期78卷 19861-19875页

作者： Fu, Zhengxin Cheng, Yuqiao Yu, Bin Zhengzhou Informat Sci & Technol Inst Zhengzhou Henan Peoples R China Zhengzhou Informat Sci & Technol Inst Dept Comp Sci & Informat Engn Zhengzhou Henan Peoples R China

Owing to efficient machine recognition, QR codes have been widely used in many aspects such as information management, electronic identity and mobile payment. Any QR code reader has access to the message contained in a QR code;therefore, in formation security becomes a major challenge to QR codes for private use. Moreover, human vision is hard to distinguish a QR code from others because its appearance looks randomly. In this paper, we propose a visual secret sharing scheme to achieve a rich QR code with three-layer information storage. In the three layers, the first one and the second one are respectively recognizable by human vision and QR code readers, while the third layer is designed for storing private data. Additionally, we also highlight the low computational complexity of secret extraction from the rich QR code.

关键词： QR code Visual secret sharing scheme Three-layer information storage low computational complexity

来源：评论

学校读者我要写书评

暂无评论

A novel highly efficient alternating direction method of multipliers for large-scale trimmed concave SVM

引用

APPLIED SOFT COMPUTING 2024年 167卷

作者： Wang, Huajun Li, Wenqian Changsha Univ Sci & Technol Dept Math & Stat Changsha Peoples R China Hunan Normal Univ Coll Life Sci Changsha Peoples R China

Support vector machine (SVM) is widely utilized for classification in diverse fields thanks to its superior performance. However, when tackling large-scale SVM problems, it encounters high computational complexity. To address this issue, we develop a novel trimmed concave loss SVM model called L tri-SVM, which achieves sparsity and robustness simultaneously. The new optimality theory for the nonconvex and nonsmooth L tri- SVM is developed by our new constructed proximal stationary point. Based on which, a new fast alternating direction method of multipliers (ADMM) with working set and low computational complexity is established for addressing L tri-SVM. Based on our new established optimality theory, our proposed algorithm will divides the training dataset into two distinct categories: the non-working and the working sets. In every training iteration, the algorithm modifies the parameters associated with the working set, while ensuring that the parameters belonging to the non-working set remain unchanged. This algorithm enables the updating of parameters on smaller datasets, which in turn decreases computational complexity and improves the efficiency of runtime. Further, we prove that our algorithm is global convergence. Numerical experiments have confirmed the excellent performance of our algorithm in terms of accuracy in classification, number of support vector and computational speed, surpassing nine other leading solvers. For instance, when solving the real dataset with over 10 7 samples, our algorithm is able to complete the task in just 18.92 s, which is much faster than other solvers that take at least 650.4 s.

关键词： low computational complexity Optimality theory Large-scale trimmed concave SVM Sparsity and robustness Global convergence

来源：评论

学校读者我要写书评

暂无评论

Efficient fast learning automata

引用

INFORMATION SCIENCES 2003年第1-4期157卷 121-133页

作者： Obaidat, MS Papadimitriou, GI Pomportsis, AS Monmouth Univ Dept Comp Sci W Long Branch NJ 07764 USA Aristotle Univ Thessaloniki Dept Informat GR-54006 Thessaloniki Greece

A new class of learning automata which are capable of supporting high-speed real-time applications is introduced. The proposed learning automata have a unique characteristic: they are capable of performing both probability updating and action selection with a computational complexity which is independent of the number of actions. Apart from their low computational complexity, the proposed automata are capable of achieving a high performance when operating in non-stationary stochastic environments. (C) 2003 Elsevier Inc. All rights reserved.

关键词： learning automata non-stationary stochastic environments high performance low computational complexity

来源：评论

学校读者我要写书评

暂无评论

A new two-level QR code with visual cryptography scheme

引用

MULTIMEDIA TOOLS AND APPLICATIONS 2018年第16期77卷 20629-20649页

作者： Cheng, Yuqiao Fu, Zhengxin Yu, Bin Shen, Gang Zhengzhou Informat Sci & Technol Inst Zhengzhou Henan Peoples R China

The Quick Response (QR) code is designed for information storage recognizable by machine vision. With the popularity of QR code applications, the security of QR codes becomes hot issues concerned by many scholars. In this paper, a novel QR code with two-level information storage is designed to protect private messages. In the meantime, the public level can be directly decoded by any standard QR reader. In contrast to other studies, the computational complexity of the proposed scheme is reduced by combining with the theory of visual cryptography scheme (VCS). In addition, an important characteristic of the QR code, error correction capability, is preserved in this paper, guaranteeing the robustness to QR code damage. Experimental results and analysis show that the proposed scheme is both feasible and reasonably secure, further enriching the application fields and outperforming the previous schemes significantly.

关键词： QR code Two-level information storage VCS low computational complexity High error correction

来源：评论

学校读者我要写书评

暂无评论

Efficient Window-Based Channel Estimation for OFDM System in Multi-Path Fast Time-Varying Channels

引用

IEICE TRANSACTIONS ON COMMUNICATIONS 2015年第11期E98B卷 2330-2340页

作者： Jan, Yih-Haw Nan Kai Univ Technol Dept Comp & Commun Engn Nantou Taiwan Nan Kai Univ Technol Dept Digital Living Innovat Nantou Taiwan

Orthogonal frequency division multiplexing (OFDM) channel estimation is the key technique used in broadband wireless networks. The Doppler frequency caused by fast mobility environments will cause inter-carrier interference (ICI) and degrade the performance of OFDM systems. Due to the severe ICI, channel estimation becomes a difficult task in higher mobility scenarios. Our aim is to propose a pilot-aided channel estimation method that is robust to high Doppler frequency with low computational complexity and pilot overheads. In this paper, the time duration of each estimate covers multiple consecutive OFDM symbols, named a "window". A close-form of polynomial channel modeling is derived. The proposed method is initialized to the least squares (LS) estimates of the channels corresponding to the time interval of the pilot symbols within the window. Then, the channel interpolation is performed in the entire window. The results of computer simulations and computation complexity evaluations show that the proposed technique is robust to high Doppler frequency with low computation complexity and low pilot overheads. Compared with the state-of-the-art method and some conventional methods, the new technique proposed here has much lower computational complexity while offering comparable performance.

关键词： OFDM low computational complexity channel estimation polynomial interpolation

来源：评论

学校读者我要写书评

暂无评论

A joint OFDM PAPR reduction and data decoding scheme with no SI estimation

引用

EURASIP JOURNAL ON WIRELESS COMMUNICATIONS AND NETWORKING 2016年第1期2016卷 1页

作者： Adegbite, Saheed A. McMeekin, Scott G. Stewart, Brian G. Glasgow Caledonian Univ Sch Engn & Built Environm 70 Cowcaddens Rd Glasgow G4 0BA Lanark Scotland

The need for side information (SI) estimation poses a major challenge when selected mapping (SLM) is implemented to reduce peak-to-average power ratio (PAPR) in orthogonal frequency division multiplexing (OFDM) systems. Recent studies on pilot-assisted SI estimation procedures suggest that it is possible to determine the SI without the need for SI transmission. However, SI estimation adds to computational complexity and implementation challenges of practical SLM-OFDM receivers. To address these technical issues, this paper presents the use of a pilot-assisted cluster-based phase modulation and demodulation procedure called embedded coded modulation (ECM). The ECM technique uses a slightly modified SLM approach to reduce PAPR and to enable data recovery with no SI transmission and no SI estimation. In the presence of some non-linear amplifier distortion, it is shown that the ECM method achieves similar data decoding performance as conventional SLM-OFDM receiver that assumed a perfectly known SI and when the SI is estimated using a frequency-domain correlation approach. However, when the number of OFDM subcarriers is small and due to the clustering in ECM, the modified SLM produces a smaller PAPR reduction gain compared with conventional SLM.

关键词： low computational complexity Orthogonal frequency division multiplexing (OFDM) Peak-to-average power ratio (PAPR) Selective mapping (SLM) Side information (SI) estimation

来源：评论

学校读者我要写书评

暂无评论

On low-complexity adaptive wireless push-based data broadcasting

引用

INTERNATIONAL JOURNAL OF COMMUNICATION SYSTEMS 2014年第1期27卷 194-200页

作者： Polatoglou, Marianna Nicopolitidis, Petros Papadimitriou, Georgios I. Aristotle Univ Thessaloniki Dept Informat Thessaloniki 54124 Greece

This letter addresses a low-complexity adaptive wireless data broadcasting system. Specifically, it proposes a way for reducing the computational complexity of the estimation process for the item demands, which in turn leads to a lower computational complexity to the broadcast server for selecting a data item to broadcast. We assume no a priori knowledge of the client demands for information items as happens in real environments. Simulation results reveal that the lowering of the computational complexity of the proposed system does not affect the performance offered to the system clients. Copyright (c) 2012 John Wiley & Sons, Ltd.

关键词： low computational complexity adaptive data broadcasting push-based systems asymmetric wireless environment

来源：评论

学校读者我要写书评

暂无评论

low-complexity Hybrid Precoding for Multi-User MmWave Systems With low-Resolution Phase Shifters

引用

IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY 2019年第10期68卷 9774-9784页

作者： Dong, Fuwang Wang, Wei Wei, Zhenyu Harbin Engn Univ Coll Automat Harbin 150001 Heilongjiang Peoples R China

Large-scale antennas systems have been attracted significant attention to provide considerable array gains for millimeter wave (mmWave) communications systems. However, huge hardware complexity and power consumptions make the conventional precoding architectures impractical to realize. Recently, hybrid precoding structures for mmWave communications systems are widely investigated to save the cost and achieve satisfactory performance. Unfortunately, most existing precoder designs lie in the assumption of infinite or high-resolution phase shifters (PSs), which are also difficult to realize in real-world. Motivated by this fact, a low complexity hybrid precoder and combiner design for multi-user systems with two-bit resolution PSs is proposed in this paper, which provides an appropriate trade-off between energy efficiency and system sum-rate. Specifically, the candidate sets for analog beamforming are constructed by introducing the auxiliary variables, in which the number of elements is the same as that of antennas. Accordingly, the computational complexity of searching optimal beamformer is significantly reduced over proposed codebooks. Then, to avoid high complexity singular value decomposition (SVD) of channel matrix, we propose a rank-1 approximate decomposition algorithm based-on principal component analysis technique. At last, the user scheduling algorithm is presented to further improve the performance in multiuser-scenario. Simulation results verified the effectiveness of the proposed algorithms, the performance of which is close to the optimal unquantized precoder.

关键词： Precoding Radio frequency complexity theory Array signal processing Baseband Interference Millimeter wave (mmWave) multi-user communications hybrid precoder and combiner multiple-input multiple-output (MIMO) low-resolution phase shifters low computational complexity

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：