Commit 73f6c2b7 authored by Konstantinos Chatzilygeroudis's avatar Konstantinos Chatzilygeroudis
Browse files

Fix for kernel docs [ci skip]

parent aedc90e1
......@@ -20,7 +20,7 @@ namespace limbo {
Exponential kernel (see :cite:`brochu2010tutorial` p. 9).
.. math::
k(v_1, v_2) = \sigma^2\exp \Big(-\frac{1}{\l^2} ||v_1 - v_2||^2\Big)
k(v_1, v_2) = \sigma^2\exp \Big(-\frac{1}{l^2} ||v_1 - v_2||^2\Big)
Parameters:
- ``double sigma_sq`` (signal variance)
......
......@@ -9,7 +9,6 @@ namespace limbo {
namespace defaults {
struct kernel_maternthreehalves {
/// @ingroup kernel_defaults
/// This is is sigma squared!
BO_PARAM(double, sigma_sq, 1);
/// @ingroup kernel_defaults
BO_PARAM(double, l, 1);
......
......@@ -6,7 +6,9 @@
namespace limbo {
namespace defaults {
struct kernel_squared_exp_ard {
/// @ingroup kernel_defaults
BO_PARAM(int, k, 0); //equivalent to the standard exp ARD
/// @ingroup kernel_defaults
BO_PARAM(double, sigma_sq, 1);
};
}
......@@ -20,9 +22,13 @@ namespace limbo {
Computes the squared exponential covariance like this:
.. math::
k_{SE}(x, y) = \alpha^2 \exp \Big(-\frac{1}{2}(x-y)^TM(x-y)\Big),
k_{SE}(x, y) = \sigma^2 \exp \Big(-\frac{1}{2}(x-y)^TM(x-y)\Big),
with :math:`M = \Lambda\Lambda^T + diag(l_1^{-2}, \dots, l_n^{-2})` being the characteristic length scales and :math:`\alpha` describing the variability of the latent function. The parameters :math:`l_1^2, \dots, l_n^2, \alpha, \Lambda` are expected in this order in the parameter array.
with :math:`M = \Lambda\Lambda^T + diag(l_1^{-2}, \dots, l_n^{-2})` being the characteristic length scales and :math:`\alpha` describing the variability of the latent function. The parameters :math:`l_1^2, \dots, l_n^2, \Lambda` are expected in this order in the parameter array. :math:`\Lambda` is a :math:`D\times k` matrix with :math:`k<D`.
Parameters:
- ``double sigma_sq`` (signal variance)
- ``int k`` (number of columns of :math:`\Lambda` matrix)
Reference: :cite:`Rasmussen2006`, p. 106 & :cite:`brochu2010tutorial`, p. 10
\endrst
......
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment