RVM_matlabToolBox
所属分类：人工智能/神经网络/深度学习
开发工具：matlab
文件大小：154KB
下载次数：1457
上传日期：2009-12-10 10:11:35
上 传 者：
BETATA
说明： 相关向量机(RVM)的matlab源程序，包含快速算法，内含代码使用说明。
RVM采取是与支持向量机相同的函数形式稀疏概率模型，对未知函数进行预测或分类。 优点：
(1) 不仅仅输出预测目标量的点估计值,还可以输出预测值的分布.
(2) 使用更少数量的支持向量,从而显著减少输出目标量预测值的计算时间.
(3) RVM不需要估计过多的参数.
(4) RVM对是否满足Mercer 定理的核函数没有限制,适应性更好.
(Relevance Vector Machine (RVM) of the matlab source code, including the fast algorithm that contains the code instructions. RVM to support vector machines with the same function form of sparse probabilistic model to predict the unknown function, or classification. Advantages: (1) The goal is not only the amount of the output forecast point estimates, but also the distribution of the output forecast. (2) use less number of support vectors, thus significantly reduce the amount of predictive value of the output goal of computing time. (3) RVM does not require too many parameters estimated. (4) RVM on whether to satisfy Mercer' s theorem is no limit on nuclear function, adaptability and better.)
文件列表:
SB2_Release_200\SB2_Release_200\licence.txt (15402, 2007-04-30)
SB2_Release_200\SB2_Release_200\SB2_ControlSettings.m (4564, 2009-03-13)
SB2_Release_200\SB2_Release_200\SB2_Diagnostic.m (3845, 2009-03-13)
SB2_Release_200\SB2_Release_200\SB2_FormatTime.m (1627, 2009-03-13)
SB2_Release_200\SB2_Release_200\SB2_FullStatistics.m (6006, 2009-03-13)
SB2_Release_200\SB2_Release_200\SB2_Initialisation.m (7449, 2009-03-13)
SB2_Release_200\SB2_Release_200\SB2_Likelihoods.m (2240, 2009-03-13)
SB2_Release_200\SB2_Release_200\SB2_Manual.pdf (133380, 2009-03-12)
SB2_Release_200\SB2_Release_200\SB2_ParameterSettings.m (3155, 2009-03-13)
SB2_Release_200\SB2_Release_200\SB2_PosteriorMode.m (6419, 2009-03-13)
SB2_Release_200\SB2_Release_200\SB2_PreProcessBasis.m (1898, 2009-03-13)
SB2_Release_200\SB2_Release_200\SB2_Sigmoid.m (1145, 2009-03-13)
SB2_Release_200\SB2_Release_200\SB2_UserOptions.m (5926, 2009-03-13)
SB2_Release_200\SB2_Release_200\SparseBayes.m (25270, 2009-03-13)
SB2_Release_200\SB2_Release_200\SparseBayesDemo.m (9814, 2009-03-13)
SB2_Release_200\SB2_Release_200 (0, 2009-04-27)
SB2_Release_200 (0, 2009-04-27)
SPARSEBAYES Matlab Toolbox
Version 2.00
-- INTRODUCTION --
"SparseBayes" is a package of Matlab functions designed to implement
an efficient learning algorithm for "Sparse Bayesian" models.
The "Version 2" package is an expanded implementation of the algorithm
detailed in:
Tipping, M. E. and A. C. Faul (2003). "Fast marginal likelihood
maximisation for sparse Bayesian models." In C. M. Bishop and
B. J. Frey (Eds.), Proceedings of the Ninth International Workshop
on Artificial Intelligence and Statistics, Key West, FL, Jan 3-6.
This paper, the accompanying code, and further information regarding
Sparse Bayesian models and the related "relevance vector machine" may
be obtained from:
http://www.relevancevector.com
-- CODE STATUS --
March 2009: The code is currently at Version 2.0, and has seen limited
testing under Matlab Version 7.4.
-- GETTING STARTED --
The SparseBayes distribution comes with a basic user manual: see
SB2_Manual.pdf.
In summary, there are a number of files supplied in the SB2
distribution, but the user should only need to call a subset of them.
To set up options and parameters:
SB2_UserOptions.m
SB2_ParameterSettings.m
The main algorithm
SparseBayes.m
Type "help SparseBayes" etc. at the Matlab prompt for further details
on how to use these functions.
There is a also a simple demonstration program, showing how
SparseBayes may be used, in:
SparseBayesDemo.m
-- LICENCE --
Note that the "SparseBayes" software is supplied subject to version 2
of the "GNU General Public License" (detailed in the file "licence.txt").
SPARSEBAYES is free software; you can redistribute it and/or modify it
under the terms of the GNU General Public License as published by the
Free Software Foundation; either version 2 of the License, or (at your
option) any later version.
SPARSEBAYES is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
General Public License for more details.
You should have received a copy of the GNU General Public License
along with SPARSEBAYES in the accompanying file "licence.txt"; if not,
write to the Free Software Foundation, Inc., 51 Franklin St, Fifth
Floor, Boston, MA 02110-1301 USA
-- ACKNOWLEDGEMENTS --
The author would like to thank Mark Hatton, Anita Faul, Ian Nabney,
Arnulf Graf and Gavin Cawley for their assistance in producing this
code.
--
Mike Tipping
www.relevancevector.com
m a i l [at] m i k e t i p p i n g . c o m
This README file (Readme.txt) was created on 13-Mar-2009 at 8:58 AM.
近期下载者：
相关文件：
收藏者：