Uncertainty estimates
The uncertainties of model parameters are derived from the scaled diagonal elements of the covariance matrix calculated by the CMA evolution strategy.
The rescaling is derived from the curvature of the cost function (at its minimum) along the line of least variance: the smallest eigenvalue of the rescaled covariance matrix is equal to the radius of the osculating circle of the cost function (at its minimum) along the direction of least variance.
Even though the rescaling is computed accurately, the covariance matrix calculated within the CMA evolution strategy is an approximation of the inverse Hessian of the cost function (at its minimum). In consequence, Especias's uncertainty and covariance estimates are approximations, too.
The figure Covariance matrix adaption in evolution strategies illustrates how the CMA evolution strategy approaches the inverse Hessian of an objective function.