Hi guys,
I am quite new to GPML and just learned the basic concepts of GPML from
Rasmussen's book and tutorial. In the document of GPML in scikit-learn, one
of its disadvantages is stated as the loss of efficiency in high
dimensional spaces. But I cannot see this point based on the the formula
which calculates posterior distribution of f:
p( f(x)|X, y,Mi) ~ GP (m_post(x) = k(x, X)[K(X, X) + sigma_noise **2 *I
]**(-1) * y,
k_post(x, x') = k(x, x') - k(x, X)[K(X, X)
+ sigma_noise**2 *I]**(-1) * k(X, x') )
The notation here is from Rasmussen's tutorial. I don't see calculation of
any term here is limited by the high dimension of features. Can anyone help
me? Thank you very much.
Cheers,
Tao
------------------------------------------------------------------------------
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here
http://p.sf.net/sfu/sfd2d-msazure
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
