# Finite Differences

We can compute the gradient $g$ using standard finite difference methods, as follows

$δθ_{j}δU (θ)=2ϵU(θ+ϵe_{j})−U(θ−ϵe_{j}) $

- This is what I saw with the Numerical Gradient, centered difference formula!

We can compute the gradient $g$ using standard finite difference methods, as follows

$δθ_{j}δU (θ)=2ϵU(θ+ϵe_{j})−U(θ−ϵe_{j}) $

- This is what I saw with the Numerical Gradient, centered difference formula!