检索结果-内蒙古大学图书馆

stochastic gradient algorithm under (h,φ)-entropy criterion

CIRCUITS SYSTEMS AND SIGNAL PROCESSING 2007年第6期26卷 941-960页

作者： Chen, B. Hu, J. Pu, L. Sun, Z. Tsinghua Univ Dept Comp Sci & Technol Beijing 100084 Peoples R China

Motivated by the work of Erdogmus and Principe, we use the error (h, phi)-entropy as the supervised adaptation criterion. Several properties of the (h, phi)-entropy criterion and the connections with traditional error criteria are investigated. By a kernel estimate approach, we obtain the nonparametric estimator of the instantaneous (h, phi)-entropy. Then, we develop the general stochastic information gradient algorithm, and derive the approximate upper bound for the step size in the adaptive linear neuron training. Moreover, the (h, phi) pair are optimized to improve the performance of the proposed algorithm. For the finite impulse response identification with white Gaussian input and noise, the exact optimum phi function is derived. Finally, simulation experiments verify the results and demonstrate the noticeable performance improvement that may be achieved by the optimum (h, phi)-entropy criterion.

关键词： (h, phi)-entropy stochastic gradient algorithm adaptive filtering kernel methods

来源：评论

学校读者我要写书评

暂无评论

stochastic gradient algorithm for state space system with d-step delay

Stochastic gradient algorithm for state space system with <i...

引用

26th Chinese Control and Decision Conference (CCDC)

作者： Gu, Ya Ding, Rui Jiangnan Univ Minist Educ Key Lab Adv Proc Control Light Ind Wuxi 214122 Peoples R China

ISBN: (纸本)9781479937066

This paper researches parameter and state estimation problems for linear systems with d-step delay. Combining the linear transformation and the property of the shift operator, the canonical state space model with d-step delay is transformed into an identification model. The stochastic gradient algorithm is raised to identify the parameter vectors. Finally, an example is presented to validate the given algorithms.

关键词： stochastic gradient algorithm State space models Time-delay

来源：评论

学校读者我要写书评

暂无评论

Further results on almost sure convergence of the stochastic gradient algorithm

引用

INTERNATIONAL JOURNAL OF ADAPTIVE CONTROL AND SIGNAL PROCESSING 1999年第3期13卷 161-181页

作者： Radenkovic, MS Univ Colorado Dept Elect Engn Denver CO 80217 USA

In this paper, we consider the rate of convergence of the parameter estimation error and the cost function for the stochastic gradient-type algorithm. The problem is solved in the case of the minimum-variance stochastic adaptive control. It is proven that the cost function has the rate of convergence, comparable with the one established for the least-squares algorithm. Comparison of the two algorithms is made under the same conditions related to the process noise and the structure of the system model. In addition it is shown, that the convergence rate for the parameter estimation error, is of the same order as in the case of extended least-squares algorithm. Copyright (C) 1999 John Wiley & Sons, Ltd.

关键词： stochastic adaptive control stochastic gradient algorithm almost sure convergence rate of convergence

来源：评论

学校读者我要写书评

暂无评论

Improved Redundant Rule-Based stochastic gradient algorithm for Time-Delayed Models Using Lasso Regression

引用

IEEE ACCESS 2022年 10卷 3336-3342页

作者： Zhao, Hangtao Lv, Lixin Ji, Yuejiang Wuxi Vocat Coll Sci & Technol Wuxi 214122 Jiangsu Peoples R China

This paper proposes an improved redundant rule based lasso regression stochastic gradient (RR-LR-SG) algorithm for time-delayed models. The improved SG algorithm can update the parameter elements with different step-sizes and directions, thus it is more adaptive;while the lasso regression method can pick out the small weights from the redundant parameter vector, it therefore can obtain the time-delay easily. To show the effectiveness of the proposed algorithm, the convergence analysis is also given. The simulated numerical results are consistent with the analytically derived results of the proposed algorithm.

关键词： Convergence Licenses Iterative methods Cost function Mathematical models Lead Costs Parameter estimation time-delayed model lasso regression redundant rule stochastic gradient algorithm

来源：评论

学校读者我要写书评

暂无评论

Bias compensated stochastic gradient algorithm for identification of an ARX-type nonlinear rational model and its application in modeling of the dynamic of the cellular toxicity

引用

INTERNATIONAL JOURNAL OF ROBUST AND NONLINEAR CONTROL 2022年第9期32卷 5268-5280页

作者： Jing, Shaoxue Kiely, Janice Luxton, Richard Huaiyin Normal Univ Sch Phys & Elect Elect Engn 111 Changjiang West Rd Huaian 223300 Jiangsu Peoples R China Univ West England Dept Engn Design & Math Bristol Avon England

Nonlinear rational model (NRM) is a generalized nonlinear model, the NARMAX model and Volterra model can be regarded as its special cases. In this article, the parameter identification of a class of nonlinear rational models is studied. Due to the coupling of the model output and the information vector, the parameter identification of the NRM is very challenging. To reduce the complexity of the identification, the stochastic gradient algorithm is used. However, the estimate given by traditional stochastic gradient algorithm is biased. To obtain unbiased estimation, the bias is calculated by using the observations and the previous estimate and then compensated to the biased estimate. A variable factor considering the estimation error is introduced to speed up the algorithm. Theoretical analysis shows that the proposed algorithm can obtain an unbiased estimate and has the complexity of O(n). The proposed algorithm is validated by a numerical example and is applied in modeling the dynamics of the cellular toxicity using the tetra-ethyl ammonium chloride. Results show that the proposed algorithm can obtain an accurate estimate for the nonlinear rational model with less computation.

关键词： adaptive modeling bias compensation nonlinear rational model parameter estimation stochastic gradient algorithm

来源：评论

学校读者我要写书评

暂无评论

A Fast stochastic gradient algorithm: Maximal Use of Sparsification Benefits under Computational Constraints

引用

IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES 2010年第2期E93A卷 467-475页

作者： Yukawa, Masahiro Utschick, Wolfgang RIKEN BSI Lab Math Neurosci Wako Saitama 3510198 Japan Tech Univ Munich Associate Inst Signal Proc D-80290 Munich Germany

In this paper, we propose a novel stochastic gradient algorithm for efficient adaptive filtering. The basic idea is to sparsity the initial error vector and maximize the benefits from the sparsification under computational constraints. To this end, we formulate the task of algorithm-design as a constrained optimization problem and derive its (non-trivial) closed-form solution. The computational constraints, are formed by focusing on the fact that the energy of the sparsified error vector concentrates at the first few components. The numerical examples demonstrate that the proposed algorithm achieves the convergence as fast as the computationally expensive method based on the optimization without the computational constraints.

关键词： adaptive filter stochastic gradient algorithm proportionate adaptive filtering

来源：评论

学校读者我要写书评

暂无评论

Global Mapping Analysis: stochastic gradient algorithm in Multidimensional Scaling

引用

IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS 2012年第2期E95D卷 596-603页

作者： Matsuda, Yoshitatsu Yamaguchi, Kazunori Aoyama Gakuin Univ Sagamihara Kanagawa 2525258 Japan Univ Tokyo Tokyo 1538902 Japan

In order to implement multidimensional scaling (MDS) efficiently, we propose a new method named "global mapping analysis" (GMA), which applies stochastic approximation to minimizing MDS criteria. GMA can solve MDS more efficiently in both the linear case (classical MDS) and non-linear one (e.g., ALSCAL) if only the MDS criteria are polynomial. GMA separates the polynomial criteria into the local factors and the global ones. Because the global factors need to be calculated only once in each iteration, GMA is of linear order in the number of objects. Numerical experiments on artificial data verify the efficiency of GMA. It is also shown that GMA can find out various interesting structures from massive document collections.

关键词： multidimensional scaling stochastic gradient algorithm text mining

来源：评论

学校读者我要写书评

暂无评论

Law of the iterated logarithm for a constant-gain linear stochastic gradient algorithm

引用

SIAM JOURNAL ON CONTROL AND OPTIMIZATION 2000年第2期39卷 533-570页

作者： Joslin, JA Heunis, AJ Siemens Canada Kanata ON K2K 2M5 Canada Univ Waterloo Dept Elect & Comp Engn Waterloo ON N2L 3G1 Canada

We study almost-sure limiting properties, taken as epsilon SE arrow 0, of the finite horizon sequence of random estimates {theta(0)(epsilon), theta(1)(epsilon), theta(2)(epsilon),..., theta([T/epsilon])(epsilon)} for the linear stochastic gradient algorithm theta(n+1)(epsilon) = theta(n)(epsilon)+epsilon [a(n+1) - (theta(n)(epsilon))' Xn+1]Xn+1, theta(o)(epsilon) (Delta)(=)theta* nonrandom, where T is an element of (0,infinity) is an arbitrary constant, epsilon is an element of (0, 1] is a (small) adaptation gain, and {a(n)} and {X-n} are data sequences which drive the algorithm. These limiting properties are expressed in the form of a functional law of the iterated logarithm.

关键词： stochastic gradient algorithm L-mixing processes functional law of the iterated logarithm

来源：评论

学校读者我要写书评

暂无评论

Prediction of the performance and exhaust emissions of a compression ignition engine using a wavelet neural network with a stochastic gradient algorithm

引用

ENERGY 2018年 142卷 1128-1138页

作者： Molkdaragh, R. Rahimi Jafarmadar, S. Khalilaria, Sh Saraee, H. Soukht Urmia Univ Mech Engn Dept Orumiyeh *** West Azerbaijan Iran

The purpose of this research is to use a wavelet neural network (WNN) and stochastic gradient algorithm (SGA) to predict the performance and exhaust emissions of a compression ignition engine with nanoparticles-diesel fuel. The percentage of the additive of nanoparticles to the fuel ranges between 20 and 80 ppm. A model of WNN has been applied in order to predict the relationship between the power, fuel consumption (FC), specific fuel consumption (SFC), CO, NOx, and HC with the amount of nano particles at different speeds. The input variables are of two parameters (the percentage of nanoparticles and engine speed), while the output variables are of six parameters (power, FC, SFC, CO, NOx, and HC). In this work, considering the characteristics of the utilized wavelet function and application of the SGA method, satisfactory results were obtained in prediction of exhaust emissions and performance of the target engine. In addition, two common artificial neural networks (ANNs) (back propagation (BP) and non-linear autoregressive with exogenous input (NARX)) were used in predicting the performance of internal combustion engines compared with WNN results. Therefore, evaluation results of these three networks showed that the WNN with the SGA are very accurate and useful method to perform the prediction and model nonlinear phenomena of internal combustion engines. (C) 2017 Elsevier Ltd. All rights reserved.

关键词： WNN BPNN NARXNN stochastic gradient algorithm Compression ignition engine Performance Emissions

来源：评论

学校读者我要写书评

暂无评论

Aitken-Based stochastic gradient algorithm for ARX Models with Time Delay

引用

CIRCUITS SYSTEMS AND SIGNAL PROCESSING 2019年第6期38卷 2863-2876页

作者： Wang, Cheng Li, Kaicheng Jiangnan Univ Sch Internet Things Engn Minist Educ Key Lab Adv Proc Control Light Ind Wuxi 214122 Peoples R China Beijing Jiaotong Univ Natl Engn Res Ctr Rail Transportat Operat & Contr Beijing 100044 Peoples R China

This paper proposes an Aitken-based stochastic gradient algorithm for ARX models with time delay. By using the redundant rule, the ARX model can be transformed into an augmented model. Based on the Aitken method, the parameters of the augmented model can be estimated, and then, the unknown parameters of the ARX model and the time delay can be computed. The performance of the Aitken-based stochastic gradient algorithm is then analyzed. Furthermore, a numerical example and a real system example are provided to show the effectiveness of the proposed algorithm. Compared with the traditional stochastic gradient algorithm, the Aitken-based stochastic gradient algorithm achieves better convergence performance that the parameter estimation errors converge below 3% in both examples after 400 steps.

关键词： Parameter estimation ARX model Time delay Aitken method stochastic gradient algorithm

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：