doc: bfgs_search_prior
Optimize prior hyper-parameter with respect to the pseudo-likelihood, using the BFGS algorithm
Syntax
HP = bfgs_search_prior(Xt, Yt, HPini, kfun, basis)
Arguments
- Xt matrix (n, d) where n is the number of data points and d is the dimension
- Yt vector (n, 1) of noisy observations
- HPini vector (h, 1) of initial hyper-parameters formatted as : [log(sf2) log(sn) log(w1) log(w2)]
- kfun kernel function such as kernel_se
- basis basis function such as basis_none
Outputs
- HP vector (h, 1) of locally optimal log hyper-parameters
- kernel found kernel function
- noise found noise standard deviation