One thing to keep in mind too is that gp.Marginal assumes that there is Gaussian noise in the observations. The difference between Marginal and Latent is that for Marginal f is integrated out (since multivariate normals are conjugate). So if there is no noise, you could consider using Latent. Assuming the noise is very very small though is effectively the same thing and will probably behave better numerically.
1 Like