◆ NewtonState()

template<typename ObjFun , typename ThetaGradFun , typename CovarianceT , typename ThetaInitializer >

stan::math::internal::NewtonState::NewtonState	(	int	theta_size,
		ObjFun &&	obj_fun,
		ThetaGradFun &&	theta_grad_f,
		CovarianceT &&	covariance,
		ThetaInitializer &&	theta_init
	)

inline

Constructs Newton state with a consistent (a_init, theta_init) pair.

When the caller supplies a non-zero theta_init, a_init = Sigma^{-1} * theta_init must be provided to maintain the invariant theta = Sigma * a.

Template Parameters

ObjFun	A callable type for the objective function
ThetaGradFun	A callable type for the theta gradient function
CovarianceT	A matrix type for the covariance (must support LLT solve)
ThetaInitializer	A type for the initial theta (e.g., Eigen vector)

Parameters

theta_size	Dimension of the latent space
obj_fun	Objective function: (a, theta) -> double
theta_grad_f	Gradient function: theta -> grad
covariance	Covariance matrix for the latent variables
a_init	Initial a value consistent with theta_init
theta_init	Initial theta value