Are the min max is equivalent to max min?

  • Thread starter Thread starter EngWiPy
  • Start date Start date
  • Tags Tags
    Equivalent Max
AI Thread Summary
The discussion clarifies that the min-max and max-min operations are not generally equivalent, illustrated with a specific example involving a function defined by the parity of the sum of indices. An important inequality is introduced, stating that the infimum of the supremum is greater than or equal to the supremum of the infimum for a real-valued function. This inequality holds true in general, and the reasoning is provided through definitions of supremum and infimum. Additionally, it is confirmed that the operations max-max and min-min are interchangeable, given that the suprema and infima are defined. The conversation emphasizes the distinctions and relationships between these mathematical concepts.
EngWiPy
Messages
1,361
Reaction score
61
Hello,

Is

\underset{i}{\min}\underset{n}{\max}=\underset{n}{\max}\underset{i}{\min}?

Thanks
 
Mathematics news on Phys.org
No. Take

x_{i,n}=\left\{\begin{array}{l} 0 ~\text{if}~i+n~\text{even}\\<br /> 1 ~\text{if}~i+n~\text{odd}\end{array}\right.
 
micromass said:
No. Take

x_{i,n}=\left\{\begin{array}{l} 0 ~\text{if}~i+n~\text{even}\\<br /> 1 ~\text{if}~i+n~\text{odd}\end{array}\right.

Yeah, right.
 
In general you do not have equality between min-max and max-min, however I couldn't resist mentioning a very neat inequality I learned a bit ago.

Given a real valued function ##P(x,y): X \times Y \rightarrow \mathbb{R}## one has

$$\inf_{x\in X} \sup_{y\in Y} P(x,y) \geq \sup_{y\in Y} \inf_{x\in X} P(x,y)$$

If you aren't familiar with sup (supremum - least upper bound) and inf (infimum - greatest lower bound) you can think of sup as max and inf as min.

I learned this from Robert McCann, it's in the following article. Skip to 2.2 Game Theory for an intuitive 'proof'.
http://www.math.toronto.edu/mccann/papers/FiveLectures.pdf
 
theorem4.5.9 said:
In general you do not have equality between min-max and max-min, however I couldn't resist mentioning a very neat inequality I learned a bit ago.

Given a real valued function ##P(x,y): X \times Y \rightarrow \mathbb{R}## one has

$$\inf_{x\in X} \sup_{y\in Y} P(x,y) \geq \sup_{y\in Y} \inf_{x\in X} P(x,y)$$

If you aren't familiar with sup (supremum - least upper bound) and inf (infimum - greatest lower bound) you can think of sup as max and inf as min.

I learned this from Robert McCann, it's in the following article. Skip to 2.2 Game Theory for an intuitive 'proof'.
http://www.math.toronto.edu/mccann/papers/FiveLectures.pdf

Is this inequality true in general?
 
S_David said:
Is this inequality true in general?
Yes. To see this pick x&#039;\in X and y&#039;\in Y. Clearly \sup_Y P(x&#039;,y)\geq \inf_X P(x,y&#039;), and since this is true for all y&#039;\in Y we have \sup_Y P(x&#039;,y)\geq \sup_Y \inf_X P(x,y), by definition of the supremum. Similarly, since this is true for all x&#039;\in X we have \inf_X\sup_Y P(x,y)\geq \sup_Y \inf_X P(x,y) by definition of the infimum.
 
Last edited:
dcpo said:
Yes. To see this pick x&#039;\in X and y&#039;\in Y. Clearly \sup_Y P(x&#039;,y)\geq \inf_X P(x,y&#039;), and since this is true for all y&#039;\in Y we have \sup_Y P(x&#039;,y)\geq \sup_Y \inf_X P(x,y), by definition of the supremum. Similarly, since this is true for all x&#039;\in X we have \inf_Y\sup_Y P(x,y)\geq \sup_Y \inf_X P(x,y) by definition of the infimum.

OK, what abou max max or min min, are they interchangeable?
 
S_David said:
OK, what abou max max or min min, are they interchangeable?
Yes. Consider \sup_X\sup_Y P(x,y). By definition we have \sup_X\sup_Y P(x,y)\geq P(x&#039;,y&#039;) for all x&#039;\in X,y&#039;\in Y, so \sup_X\sup_Y P(x,y)\geq \sup_{X\times Y} P(x,y) by definition of \sup. Conversely, if \sup_X\sup_Y P(x,y)\not\leq \sup_{X\times Y} P(x,y) then there is x&#039;\in X with \sup_Y P(x&#039;,y)\not\leq \sup_{X\times Y} P(x,y), and thus there is y&#039;\in Y with \sup_Y P(x&#039;,y&#039;)\not\leq \sup_{X\times Y} P(x,y), which would be a contradiction. So \sup_X\sup_Y P(x,y)= \sup_{X\times Y} P(x,y) and a symmetry argument gives \sup_X\sup_Y P(x,y)= \sup_{X\times Y} P(x,y)=\sup_Y\sup_X P(x,y).

Edit: We require that the suprema and infima are defined, I forgot to mention that in my original response.
 
dcpo said:
Yes. Consider \sup_X\sup_Y P(x,y). By definition we have \sup_X\sup_Y P(x,y)\geq P(x&#039;,y&#039;) for all x&#039;\in X,y&#039;\in Y, so \sup_X\sup_Y P(x,y)\geq \sup_{X\times Y} P(x,y) by definition of \sup. Conversely, if \sup_X\sup_Y P(x,y)\not\leq \sup_{X\times Y} P(x,y) then there is x&#039;\in X with \sup_Y P(x&#039;,y)\not\leq \sup_{X\times Y} P(x,y), and thus there is y&#039;\in Y with \sup_Y P(x&#039;,y&#039;)\not\leq \sup_{X\times Y} P(x,y), which would be a contradiction. So \sup_X\sup_Y P(x,y)= \sup_{X\times Y} P(x,y) and a symmetry argument gives \sup_X\sup_Y P(x,y)= \sup_{X\times Y} P(x,y)=\sup_Y\sup_X P(x,y).

Edit: We require that the suprema and infima are defined, I forgot to mention that in my original response.

Yes, of course they are defined.

Thanks
 
Back
Top