The formal definition of Fischer Information is: "For random variable X, with a likelihood function $ L(θ,X) $ and score function(with respect to parameter θ) $ s(θ;X) = \nabla [ln(L(θ,X))] $
