Concept
Given data
define residual function:
objective function:
gradient vector:
where is Jacobian matrix of .
Hessian matrix:
However itβs expensive to expend such sum and since it multiple to which is very small that we can just ignore this whole sum term.