Prove that a function from [0,1] to [0,1] is a homeomorphism

In summary, we have sets X and Y equipped with the discrete topology, and a function h from X to Y such that for each element in Y, its inverse image under h is open in X. Therefore, h is continuous. Furthermore, every element in Y has a preimage in X, making h onto. Additionally, every element in Y has a unique preimage in X, making h one-to-one. And finally, in the limit as n goes to infinity, X and Y are sets equipped with the finest topology on [0,1], and all the previous claims still hold. Therefore, h is a homeomorphism from [0,1] to [0,1].
  • #1
docnet
799
487
Homework Statement
prove that a function from [0,1] to [0,1] is a homeomorphism
Relevant Equations
prove that a function from [0,1] to [0,1] is a homeomorphism
Screen Shot 2021-10-20 at 5.01.20 PM.png


let ##X=\{0,p1,p_2,...,p_n,1\}## and ##Y=\{0,p1,p_2,...,p_n,1\}## be sets equipped with the discrete topology.

for each ##q_i## in ##Y##, the inverse image ##h^{-1}(q_i)=p_i## is open in ##X## w.r.t. to the discrete topology, so h is continuous.

every element y in Y has a preimage x in X, so h is onto.

every element y in Y has a unique preimage x in X. so ##h## is one-to one.

h is continuous, onto, and one to one so it is a homeomorphism from X to Y.

(unsure about this step)
in the limit as n goes to infinity, ##X=Y=[0,1] ## are sets equipped with the finest topology on [0,1] and all the previous claims hold. so ##h## is a homeomorphism from [0,1] to [0,1].
 
Physics news on Phys.org
  • #2
I think that you are proving the wrong thing. The problem asks for you to prove that such a function,h, exists. You should either construct such a function or you should use some existence theorem to show that there is such a function.
 
  • #3
FactChecker said:
I think that you are proving the wrong thing. The problem asks for you to prove that such a function,h, exists. You should either construct such a function or you should use some existence theorem to show that there is such a function.
okay, i think I was supposed to construct a quotient space that is equal to [0,1] and show that h is continuous on [0,1] ?
 
  • #4
I would construct an example for ##n=1##. Say ##p_1\leq q_1.## Then you can linearly stretch ##[0,p_1]## to ##[0,q_1]## and linearly compress ##[p_1,1]## to ##[q_1,1].## Since both functions have the same value ##y=q_1## at ##x=p_1## you can concatenate them. The rest goes on by induction.
 
  • Love
Likes docnet
  • #5
docnet said:
okay, i think I was supposed to construct a quotient space that is equal to [0,1] and show that h is continuous on [0,1] ?
It says to "Prove ... there is a homeomorphism, h". So you need to either construct h or use some other existence theorem to show that h exists.
 
  • Like
Likes docnet
  • #6
Let ##h(t)=q_{i−1}+(\frac{q_i−q_{i−1}}{p_i−p_{i−1}})(t−p_{i−1})##, a linear mapping between the intervals ##[p_i, p_{i+1}]##

required conditions: ##h(0)=0, h(1)=1##, and ##h(p_i)=q_i##/

claim: h satisfies the required conditions for any n.

let n=1, then ##h(p_1)=q_1##. suppose h satisfies the condition for n=k, i.e., ##h(p_i)=q_i## for values of i in ##\{0,1,...,k\}##.Then for ##n=k+1##, ##h(p_i)=q_i## for values of ## i \in \{0,1,...,k,k+1\}## so h satisfies the conditions for any n.

construct concatenations using the natural maps

##f:X/(p_i\sim p_i)\rightarrow [0,1]##
##g:Y/(q_i\sim q_i)\rightarrow [0,1]##

h is continuous because for all open sets in ##Y/(q_i\sim q_i)##, the inverse image is an open set in ##X/(p_i\sim p_i)##.
the inverse of h is continuous because for all open sets in ##X/(p_i\sim p_i)##, there is an open set in ##Y/(q_i\sim q_i)##.
h is onto and one-to one because ##Y/(q_i\sim q_i)## is covered once by the image of ##X/(p_i\sim p_i)##.

so h is a homeomorphism.
 
Last edited:

Similar threads

Back
Top