How to find hyperparameters (in Gaussian Process) by log marginal likelihood.
Suppose we have mean function as $m=ax^2 +bx + c$.
so we have $hyperparameters = \{a,b,c,sigma_y,sigma_n,l\}$;
How we calculate all these parameters by log marginal likelihood and by partial derivatives?

  • 1
    $\begingroup$ What research have you done? Where have you looked? $\endgroup$ – D.W. Oct 30 '13 at 17:25
  • $\begingroup$ I have researched previous journals for real estate and find these problem. I have found one scholar article by: Gaussian Process in Machine Learning by Carl Edward Rasmussen In this paper it is mentioned above problem. I am not able to find out how to calculate those hyperparamters practically (in Matlab). $\endgroup$ – user2907251 Nov 1 '13 at 10:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.