Consider the following program:
subject to g(x,y)=x^2+y^2-1=0
Solution. Let us denote the gradient vector of the function f(x,y) by Df(x,y). We rewrite the original program as follows.
Since DF(x,y)=(-2x-4y,-4x-2y)', Dg(x,y)=(2x,2y)', by K-T ...
A function is maximized using Kuhn-Tucker condition.