Feasible set of linear optimization problem

$\begingroup$

Consider the set

$$\{ x \in \mathbb{R}^n \mid x_1 = \cdots = x_{n-1}=0, 0 \le x_n \le 1 \}$$

Could this be the feasible set of a problem in standard form?

I'm leaning towards no, but I am unsure how to prove this.

$\endgroup$ 1

2 Answers

$\begingroup$

I'm assuming that the standard form of a constraint set is a collection of inequalities $a_1^Tx \le c_1, ..., a_m^Tx \le c_m$. In this case, yes, this set can be written in standard form.

The constraint $x_i = 0$ can be written as the equation $e_i^Tx = 0$ where $e_i$ is the $i^{th}$ standard basis vector, $e_1 = (0, ..., 1, ..., 0)^T$ where the $1$ is in the $i^{th}$ position. An equation can always be written as two inequalities, $e_i^Tx \ge 0$, $e_i^Tx \le 0$, or equivalently $-e_i^Tx \le 0, e_i^Tx \le 0.$

For the last inequalities: they are $e_n^T \ge 0$ and $e_n^T \le 1$, or equivalently $-e_n^T \le 0$ and $e_n^T \le 1.$

$\endgroup$ $\begingroup$

No.

Assume $Q$ can be represented in $\{x \mid Ax = b, x \geq 0\}$. We know $\boldsymbol{0} \in Q$, thus $b=\boldsymbol{0}$.

Let $A = [A_1, \dots, A_n]$, then $Q = \{x \mid x_1 A_1 + \dots + x_n A_n = \boldsymbol{0}, x \geq 0\}$. Since $[0, \dots, 1] \in Q$, $\;A_n = \boldsymbol{0}$ must hold.

However, in this situation, $[0, \dots, 2] \in Q$, which is a contradiction.

$\endgroup$

Your Answer

Sign up or log in

Sign up using Google Sign up using Facebook Sign up using Email and Password

Post as a guest

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

You Might Also Like