Kannan mapping and quasi-nonexpansive mapping

  • Context: MHB 
  • Thread starter Thread starter ozkan12
  • Start date Start date
  • Tags Tags
    Mapping
Click For Summary

Discussion Overview

The discussion revolves around the definitions and implications of Kannan mappings and quasi-nonexpansive mappings within the context of metric spaces. Participants explore the logical relationships between inequalities in these definitions and the implications of using strict versus non-strict inequalities.

Discussion Character

  • Technical explanation
  • Debate/contested
  • Mathematical reasoning

Main Points Raised

  • Post 1 presents the definitions of Kannan mapping and quasi-nonexpansive mapping, highlighting a potential issue with the inequality used in the proof that every Kannan mapping is quasi-nonexpansive.
  • Post 2 asserts that the strict inequality "$<$" implies the non-strict inequality "$\le$," suggesting that the last line of the proof could use "$\le$" instead of "$<$."
  • Post 3 questions the logical basis for "$< \implies \le$," seeking clarification on how this implication holds in the context of the inequalities presented.
  • Post 4 explains the logical principle behind the implication, referencing the "Disjunction Introduction Rule" and providing a reasoning framework for why "$< \implies \le$" is valid.
  • Post 5 agrees with the use of "$\le$" instead of "$<$" and inquires about the implications of substituting $x=p$ in the inequalities.
  • Post 6 challenges the earlier statements by suggesting that the relationship between "$\le$" and "$<$" is not as straightforward as presented.
  • Post 7 clarifies that while "$x \le y$" allows for either "$x < y$" or "$x = y$," it does not specify which condition holds, emphasizing the uncertainty in the relationship.
  • Post 8 acknowledges the truth of the previous statement but expresses confusion regarding the initial post's argument.

Areas of Agreement / Disagreement

Participants express differing views on the implications of the inequalities involved in the definitions, with no consensus reached on the logical connections or the appropriateness of using "$\le$" versus "$<$" in the context of Kannan mappings and quasi-nonexpansive mappings.

Contextual Notes

The discussion highlights potential ambiguities in the definitions and implications of mathematical properties, particularly regarding the use of strict and non-strict inequalities, which may depend on the specific context or assumptions made by participants.

ozkan12
Messages
145
Reaction score
0
Definition of Kannan Mapping

Let (X,d) be a complete metric space...İf for each $x,y\in X$ following condition holds, then $T:X\to X$ is Kannan mapping

$d(Tx,Ty)\le\alpha\left[d\left(x,Tx\right),d(\left(y,Ty\right)\right)]$ $\alpha\in[0,\frac{1}{2})$

Definition of Quasi-Nonexpansive Mapping

An operator $T:X\to X$ is said to be quasi nonexpansive if T has at least one fixed point in X and, for each fixed point p, we have

$d(Tx,p)\le d(x,p)$ for each x $\in$ X.

İf T is a Kannan mapping, then T is a quasi nonexpansive operator.Proof:

İndeed, if T is a Kannan operator, then from definition of Kannan mapping with y=p in set of fixed point of T we get

$d(Tx,p)\le\alpha d\left(x,Tx\right)\le\alpha\left[d(x,p)+d(p,Tx)\right]$ and hence

$d\left(Tx,p\right)\le\frac{\alpha}{1-\alpha}d\left(x,p\right)<d(x,p)$İn there we get d(Tx,p)<d(x,p)...But in definition of quasi nonexpansive operator we use "$\le$"...But in last step we use "<" İn this case how we say that every kannan mapping is a quasi non expansive operator
 
Physics news on Phys.org
$< \; \implies \; \le$. That is, $<$ is a stronger condition than $\le$, and implies it. You could just as well have used the weaker $\le$ in the very last line.
 
Dear Ackbach,

First of all, Thank you for your attention...

How "$<$" implies $\le$ ? İn nigt, I thougt that how we write this inequeality...Then I found some articles...İf we take p=x in

$$d(Tx,p)<d(x,p)$$ then we get $d(Tx,p)=d(x,p)=0$...So, we get $d(Tx,p)\le d(x,p)$...

But, how "$<\implies \le$" ? I didnt understand...as you say I can use " $\le$" in last inequeality...That is, I can write $d\left(Tx,p\right)\le d\left(x,p\right)$...İs this true ? Thank you for your attention..$<\implies \le$
 
Last edited:
This is a property of logic. In Natural Deduction, we call it the "Disjunction Introduction Rule". Copi calls it something else. It goes like this:

A implies A or B. If A is true, then A or B is true.

If $x<y$ is true, then it is also true that either $x<y$ or $x=y$. That is, if $x<y$ is true, then $x\le y$. Therefore, $x<y \; \implies \; x\le y$. So, my shorthand notation there, $< \; \implies \; \le$ is what I meant.
 
Dear Ackbach,

That is, we can use $\le$ instead of $<$ in last inequality...

Also, can we say that if we take x=p, we will obtain last inequality ? Thank you for your attention...
 
Dear Ackbach

İf $x\le y$ then we can say $x<y$ or $x=y$..İn my research it is not as you say...
 
If $x\le y$, then you can say $x<y$ or $x=y$, but you don't know which. It could be either. You definitely CANNOT narrow it down to only one of those two possibilities.
 
yes, this is true...but we don't say if $x<y$ then $x<y$ or $x=y$ ...And still in first post is strange for me ...
 

Similar threads

  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 0 ·
Replies
0
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 5 ·
Replies
5
Views
1K
  • · Replies 10 ·
Replies
10
Views
3K
  • · Replies 0 ·
Replies
0
Views
618
Replies
8
Views
3K
  • · Replies 2 ·
Replies
2
Views
1K
  • · Replies 2 ·
Replies
2
Views
2K