MHB Proving Convex Set Properties to Showing the Convexity of X-Y

Click For Summary
The discussion focuses on proving that the set difference of two convex sets, defined as Z = {x - y | x ∈ X, y ∈ Y}, is also convex. The initial approach involved assuming a point r in the segment between two points p and q in Z and attempting to show r is not in Z, which was deemed unnecessary. Instead, it was clarified that one can directly express r as a combination of points from X and Y, confirming that r remains in Z. The participants emphasized that the convexity of Z can be established without needing to assume X and Y are vector spaces, as their convexity suffices for the proof. The conversation concluded with a reminder that the properties of convex sets alone are adequate for this proof.
rputra
Messages
35
Reaction score
0
I need help on this problem:

If $X$ and $Y$ are convex sets, show that $X-Y = Z = \{x-y \mid x \in X, y \in Y\}$ is also convex.

Here are the steps I have gone so far:

Let $p \in Z$ such that $p = x_1 - y_1$, and let $q \in Z$ such that $q = x_2 - y_2$. Assume that $r$ lays in the segment line of $\overline {pq}$ so that there exists $k$ and $k'$, with $k, k' \geq 0$ and $k+k' = 1$, such that $r = kp + (1-k)q$. By way of contradiction, we are going to show that $r \notin Z$, thus implying that $Z$ is indeed concave.

Then,
$$\begin{align}
r &= kp + (1-k)q \\
&= k(x_1 - y_1) + (1-k)(x_2 - y_2) \\
&= x_1k - y_1k + x_2 - y_2 - x_2k + y_2k \\
&= \ldots \\
&= \ldots
\end{align}$$

But then I do not how to follow up. Any help or hint would be very much appreciated. Thank you for your time.
 
Last edited:
Physics news on Phys.org
Tarrant said:
I need help on this problem:

If $X$ and $Y$ are convex sets, show that $X-Y = Z = \{x-y \mid x \in X, y \in Y\}$ is also convex.

Here are the steps I have gone so far:

Let $p \in Z$ such that $p = x_1 - y_1$, and let $q \in Z$ such that $q = x_2 - y_2$. Assume that $r$ lays in the segment line of $\overline {pq}$ so that there exists $k$ and $k'$, with $k, k' \geq 0$ and $k+k' = 1$, such that $r = kp + (1-k)q$. By way of contradiction, we are going to show that $r \notin Z$, thus implying that $Z$ is indeed concave.

Then,
$$\begin{align}
r &= kp + (1-k)q \\
&= k(x_1 - y_1) + (1-k)(x_2 - y_2) \\
&= x_1k - y_1k + x_2 - y_2 - x_2k + y_2k \\
&= \ldots \\
&= \ldots
\end{align}$$

But then I do not how to follow up. Any help or hint would be very much appreciated. Thank you for your time.

I am not sure why are you invoking the contradiction approach. It's rather straightforward. And you have to show that $Z$ is convex and not concave!

Here is what you do

$$\begin{align}
r &= kp + (1-k)q \\
&= k(x_1 - y_1) + (1-k)(x_2 - y_2) \\
&= [kx_1 +(1-k)x_2 ]- [ky_1+ (1-k)y_2] \\
&= x-y
\end{align}$$
where $x=kx_1+(1-k)x_2$ and $y=ky_1+(1-k)y_2$. Note that $x\in X$ and $y\in Y$ since $X$ and $Y$ are convex. Thus $kp+(1-k)q$ lies in $X-Z=Y$, showing that $Z$ is convex.
 
caffeinemachine said:
I am not sure why are you invoking the contradiction approach. It's rather straightforward. And you have to show that $Z$ is convex and not concave!

Here is what you do

$$\begin{align}
r &= kp + (1-k)q \\
&= k(x_1 - y_1) + (1-k)(x_2 - y_2) \\
&= [kx_1 +(1-k)x_2 ]- [ky_1+ (1-k)y_2] \\
&= x-y
\end{align}$$
where $x=kx_1+(1-k)x_2$ and $y=ky_1+(1-k)y_2$. Note that $x\in X$ and $y\in Y$ since $X$ and $Y$ are convex. Thus $kp+(1-k)q$ lies in $X-Z=Y$, showing that $Z$ is convex.

My bad, my bad! Sorry for the dumb opinion. Thank you for pointing that out! Just to let you know that I forgot to mention that $X$ and $Y$ are in vector space. (Or perhaps you have already known it by implication?) That means that $X$ and $Y$ are closed under addition and scalar multiplication, am I correct in these? Thanks again and again for your time and effort.
 
Tarrant said:
My bad, my bad! Sorry for the dumb opinion. Thank you for pointing that out! Just to let you know that I forgot to mention that $X$ and $Y$ are in vector space. (Or perhaps you have already known it by implication?) That means that $X$ and $Y$ are closed under addition and scalar multiplication, am I correct in these? Thanks again and again for your time and effort.
The implication that $X-Y$ is convex follows from the convexity of $X$ and $Y$ alone. We do not need to assume that $X$ and $Y$ are vector spaces. If something is confusing you then you may start from the beginning in the next post.
 
Thread 'How to define a vector field?'
Hello! In one book I saw that function ##V## of 3 variables ##V_x, V_y, V_z## (vector field in 3D) can be decomposed in a Taylor series without higher-order terms (partial derivative of second power and higher) at point ##(0,0,0)## such way: I think so: higher-order terms can be neglected because partial derivative of second power and higher are equal to 0. Is this true? And how to define vector field correctly for this case? (In the book I found nothing and my attempt was wrong...

Similar threads

  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 25 ·
Replies
25
Views
3K
  • · Replies 6 ·
Replies
6
Views
2K
Replies
2
Views
2K
  • · Replies 40 ·
2
Replies
40
Views
3K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 34 ·
2
Replies
34
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 11 ·
Replies
11
Views
2K