Help with rewriting a compound inequality

  • Context: Undergrad 
  • Thread starter Thread starter Andrea94
  • Start date Start date
  • Tags Tags
    Inequality Optimization
Click For Summary

Discussion Overview

The discussion revolves around rewriting a compound inequality and understanding the implications of integer-convex functions in relation to inequalities. Participants explore various approaches to manipulate the inequalities and clarify the concept of integer-convexity.

Discussion Character

  • Exploratory
  • Technical explanation
  • Conceptual clarification
  • Debate/contested

Main Points Raised

  • Some participants suggest breaking down the compound inequality into two separate inequalities for easier manipulation.
  • There is a question regarding the meaning of integer-convexity and its relation to the expression $$g(k+1) + g(k-1) - 2g(k)$$.
  • One participant notes that the expression can be rewritten as $$\Delta g(k) - \Delta g(k-1)$$, but its relevance is questioned.
  • Another participant points out that for an integer-convex function, the sign of $$\Delta g(k) - \Delta g(k-1)$$ can vary, leading to uncertainty about its implications.
  • There is discussion about the conditions under which $$g(k+1) + g(k-1) - 2g(k)$$ is positive, particularly in the context of optimal values of $$k$$.
  • Some participants express confusion about the introduction of the term 'optimal' and its relevance to the current discussion.
  • One participant confirms that for a convex function, the second derivative is always positive, which relates to the expression in question.

Areas of Agreement / Disagreement

Participants express differing views on the implications of integer-convexity and the relevance of optimal values, indicating that multiple competing views remain. The discussion does not reach a consensus on the interpretation of the expressions involved.

Contextual Notes

There are limitations in the discussion regarding the assumptions made about the context of 'optimal' and the definitions of integer-convexity, which are not fully resolved.

Andrea94
Messages
21
Reaction score
8
TL;DR
Help with rewriting optimality conditions for integer-convex functions.
See attached screenshot.
Stumped on this, I'll take anything at this point (hints, solution, etc).

help.png
 
  • Like
Likes   Reactions: PhDeezNutz
Mathematics news on Phys.org
Big problems are to be split into smaller problems. You have two inequalities. Write the first one down and manipulate ! Then idem number two.

##\ ##
 
  • Like
Likes   Reactions: Andrea94
And: what does integer-convex mean for e.g. $$g(k+1) + g(k-1) - 2g(k) \quad \textsf{?} $$

##\ ##
 
BvU said:
Big problems are to be split into smaller problems. You have two inequalities. Write the first one down and manipulate ! Then idem number two.

##\ ##
Great, thanks! Didn't even think about the fact that I could do each inequality separately.
 
BvU said:
And: what does integer-convex mean for e.g. $$g(k+1) + g(k-1) - 2g(k) \quad \textsf{?} $$

##\ ##
What do you mean? The expression you've written is the same thing as $$\Delta g(k) - \Delta g(k-1)$$ but I'm not sure how that is relevant.
 
Andrea94 said:
What do you mean? The expression you've written is the same thing as $$\Delta g(k) - \Delta g(k-1)$$ but I'm not sure how that is relevant.
Right, but ##\Delta g(k) - \Delta g(k-1)## can be ##\ge 0## or ##\le 0## for an integer-convex function ... whereas ... :wink:

##\ ##
 
BvU said:
Right, but ##\Delta g(k) - \Delta g(k-1)## can be ##\ge 0## or ##\le 0## for an integer-convex function ... whereas ... :wink:

##\ ##
Still not sure what I'm supposed to see here 😅
 
The problem with inequalities is that you can only multiply (or divide) left and right with something positive. As it happens, for integer-convex functions the sign of ##g(k+1) + g(k-1) - 2g(k)## is ...

##\ ##
 
BvU said:
The problem with inequalities is that you can only multiply (or divide) left and right with something positive. As it happens, for integer-convex functions the sign of ##g(k+1) + g(k-1) - 2g(k)## is ...

##\ ##
Hm, the only thing I can think of is that if ##k## is optimal and nonzero, then ##g(k+1) + g(k-1) - 2g(k)## is always positive since for optimal ##k## we have ##\Delta g(k) > 0## and ##\Delta g(k-1) < 0##. Is this what you mean?
 
  • #10
Not sure where your 'optimal' comes from (it seems to live in your context, but not in the context of this thread ?).

But: yes, for a convex function the second derivative is always positive and so is this ##g(k+1) + g(k-1) - 2g(k)##.

I figured it might help in manipulating the inequalities ...

##\ ##
 
  • Like
Likes   Reactions: Andrea94
  • #11
BvU said:
Not sure where your 'optimal' comes from (it seems to live in your context, but not in the context of this thread ?).

But: yes, for a convex function the second derivative is always positive and so is this ##g(k+1) + g(k-1) - 2g(k)##.

I figured it might help in manipulating the inequalities ...

##\ ##
Ohh I see, thanks a lot for the help!
 

Similar threads

  • · Replies 26 ·
Replies
26
Views
2K
  • · Replies 6 ·
Replies
6
Views
3K
Replies
2
Views
1K
Replies
7
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
Replies
1
Views
578
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 11 ·
Replies
11
Views
2K
Replies
4
Views
3K
  • · Replies 3 ·
Replies
3
Views
3K