approx_for_gpc
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:839KB
下载次数:10
上传日期:2014-12-21 22:51:31
上 传 者:
2319723
说明: 高斯过程二分类中,后验概率的几种估计算法
(Approximations for Binary Gaussian Process Classifi cation)
文件列表:
approx\approxEC.m (3552, 2009-02-02)
approx\approxEP.m (5259, 2007-07-30)
approx\approxFV.m (2234, 2008-09-04)
approx\approxIVM.m (3005, 2007-04-13)
approx\approxKL.m (7876, 2008-03-06)
approx\approxLA.m (3094, 2007-07-17)
approx\approxLR.m (1835, 2008-11-04)
approx\approxOLEP.m (9158, 2008-11-04)
approx\approxSO.m (5865, 2007-04-11)
approx\approxTAP.m (1616, 2008-02-26)
approx\approxTAPnaive.m (1571, 2008-02-26)
approx\approxVB.m (9412, 2007-05-03)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\alg21.gif (31996, 2006-03-07)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\alg31.gif (43172, 2006-03-07)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\alg32.gif (34064, 2006-03-07)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\alg35.gif (66673, 2006-03-07)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\alg36.gif (32226, 2006-03-07)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\alg51.gif (60084, 2006-03-07)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\alg52.gif (41085, 2006-03-07)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\classification.html (34439, 2006-03-26)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\fig2d.gif (36772, 2006-03-15)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\fig2de1.gif (35364, 2006-03-15)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\fig2de2.gif (42683, 2006-03-15)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\fig2de3.gif (38483, 2006-03-15)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\fig2dl1.gif (34059, 2006-03-15)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\fig2dl2.gif (40431, 2006-03-15)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\fig2dl3.gif (38375, 2006-03-15)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\figepp.gif (17875, 2006-03-10)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\figepp2.gif (23299, 2006-03-10)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\figl.gif (6109, 2006-03-27)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\figl1.gif (21386, 2006-03-07)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\figlapp.gif (17854, 2006-03-09)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\figlapp2.gif (25557, 2006-03-09)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\figlf.gif (16078, 2006-03-27)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\figlm.gif (16712, 2006-03-27)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\index.html (3461, 2007-06-26)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\regression.html (16598, 2006-03-29)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\sparse-approx.html (7738, 2006-03-29)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\doc\style.css (77, 2006-03-07)
approx\gpml-matlab-v2.0-2007-06-25\gpml-matlab\gpml\approxEP.m (5095, 2007-06-27)
... ...
In the Gaussian process code [1] of Carl Edward Rasmussen, you find two approxi-
mation methods, namely Laplace's Approximation (approxLA.m) and Expectation
Propagation (approxEP.m). Examples and documentation is provided there.
In the paper, Approximations for binary Gaussian process classification, we
compare and analyze several other non-sparse methods:
- Direct Kullback-Leibler divergence minimization (approxKL.m)
- Individual variational site bounding (approxVB.m)
- Factorial Variational approximation (approxFV.m)
- Label Regression (approxLR.m)
The ADATAP approach (approxTAP.m) yields the same solution as EP but at higher
computational cost; the implementation is numerically not favorable and only
included for reasons of completeness. A simpler variant (approxTAPnaive.m) is
also provided.
Recently, we became aware of a fast single loop algorithm (approxEC.m) finding
the same solution as EP but faster.
Three sparse/online methods are included. This code is not discussed in the
paper but yields similar performance. It is primarily included to show that the
interface of the GPML code treats sparse methods as well. The code is tested
but far less reliable than the working horses (approxLA.m, approxEP.m).
- Informative Vector Machine (approxIVM.m)
- Sparse Online Gaussian Processes (approxSO.m)
- Online-EP (approxOLEP,m)
Hannes Nickisch, February 2, 2009
[1] http://www.gaussianprocess.org/gpml/code/matlab/doc/classification.html
[2] http://jmlr.csail.mit.edu/papers/volume9/nickisch08a/nickisch08a.pdf
近期下载者:
相关文件:
收藏者: