Now run your Newton's method code from part $($c$)$ again, this time initializing at the point w$\backslash$deg $=4-$ INx$1.$ While this initialization is further away from the unique minimum of g $($w$)$ than the one used in part $($c$)$ your Newton's method algorithm should converge faster starting at this point. At first glance this result seems very counterintuitive, as we $($rightfully$)$ expect that an initial point closer to a minimum will provoke more rapid convergence of Newton's method! Explain why this result actually makes sense for the particular function g $($w$)$ we are minimizing here.