![]() |
Stan Math Library
5.1.0
Automatic Differentiation
|
|
inline |
The derivative of the log likelihood wrt theta evaluated at the mode.
Compute $s_2 = \Delta_{\theta} log \pi_G(y|\phi,\eta) = -\frac{1}{2} trace((K^{-1}+W)^{-1})$
Args contains var types then their adjoints will be calculated as a side effect. | F | A functor with opertor()(Args&&...) returning a scalar |
| Theta | An Eigen Matrix |
| AMat | An Eigen Matrix |
| Stream | Type of stream for messages. |
| Args | Type of variadic arguments for likelihood function. |
| f | Log likelihood function. |
| theta | Latent Gaussian variable. |
| A | Matrix storing initial tangents for higher-order differentiation (line 21 in Algorithm 4, https://arxiv.org/pdf/2306.14976) |
| hessian_block_size | If the Hessian of the log likelihood w.r.t theta is block diagonal, size of each block. |
| msgs | Stream for messages. |
| args | Variational arguments for likelihood function. |
Definition at line 212 of file laplace_likelihood.hpp.