Unique lifting property

From Maths
Jump to: navigation, search

Statement

Let (X,J) be a topological space, suppose that (E,H) is a covering space (with covering map p:EX). Suppose (Y,K) is a connected topological space and f:YX is a continuous map, then[1]Partial:[2] - [Note 1]

  • given two lifts of f through p, say g,h:YE we have:
    • (yY[g(y)=h(y))( yY[g(y)=h(y)]i.e. that g=h )
      • In words: if there exists a point on which g and h agree then g and h are equal as functions

Bonus corollary

Recall that a logical implication is logically equivalent to the contrapositive, that is

  • (AB)(¬B¬A)

So, should the above claim be true, we also get:

  • (yY[g(y)h(y)])(yY[g(y)h(y)])
    • In words: if there exists a y0Y such that g and h disagree at y0 then they disagree everywhere.
      • Caveat:This does not mean g(Y)h(Y)= necessarily!

Proof

Grade: C
This page requires some work to be carried out
Some aspect of this page is incomplete and work is required to finish it
The message provided is:
Warning:What follows is VERY messy. I was distracted when writing it and kept switching between it and something else. However the written proof was BARELY legible so this is better than nothing. Hopefully

Warning:That grade doesn't exist!
TODO: It'd be good to put some reasoning here. Like given a zY to show g and h agree there we really need to work in terms of open neighbourhoods to z, which we can get from somewhere they're known to agree on. However this only works if it's connected, as if z is in another connected components "stuff just doesn't work" <--make this more formal


Let us make the following definitions:

  • S:={yY | h(y)=g(y)} (when we introduce the hypothesis, by that hypothesis this will be non-empty)
  • T:={yY | h(y)g(y)}

Lemma 1: S and T are disjoint.

  • Suppose ST, clearly this iffzST
    • Suppose there is such a zST, by definition of intersection that means zY[zSzT]
      • zSg(z)=h(z) and zTg(z)h(z)
        • We see g(z)=h(z)g(z) so g(z)g(z) - a contradiction!
  • We see there cannot be any elements in ST for if there was we have a contradiction. As shown
  • So ST= - i.e. S and T are disjoint


Lemma 2: ST=Y AND S=YT (and T=YS) in some form. We need S=YT and T=YS at least!

TODO: Todo


As Y is a connected topological space we see that the only sets that are both open and closed are Y itself and , if the result holds (which we very much hope it does) then S=Y and T=, so if we show S is open. As T=YS we would see T is closed as a result.

If we show T is open too, then S would be closed, thus showing they're both open and closed!

So we must have S=Y (as S is non-empty by hypothesis) and T=, but if S=Y then they agree everywhere!

TODO: Notes:Covering spaces shows it better

Proof that S is open

  • Let sS be given. There is at least one to give by hypothesis. We must find a neighbourhood of s that is fully contained in S, recall that a set is open if and only if it is a neighbourhood to all of its points, this is what we hope to show.
    • Define r:=h(s)=g(s) and notice rE
      • Define z:=p(r) and notice p(r)X
        • As zX and p is a covering map there exists an open neighbourhood, UJ of z such that U is evenly covered by p
        • Choose U to be such an open neighbourhood
          • By choice of U we see (Vα)αIH
            • such that:
              1. p1(U)=αIVα
              2. The (Vα)α are open (given by being in H)
              3. The (Vα)α are pairwise disjoint
              4. For each Vβ(Vα)α we have Vβp|ImVβU, that is the covering map restricted to its image on Vβ is a homeomorphism onto U
          • Choose (Vα)αIH to be this family of sheets of the covering
            • As zU we see p1(z)p1(U)
            • As z:=p(r) we see that rp1(z) so rp1(U)
            • As p1(U)=αIVα we see rαIVα, by definition of union:
              • (rαIVα)(βI[rVβ)
            • So βI[rVβ]
              • Define V:=Vβ where Vβ is the element of (Vα)αI with rVβ as discussed above
                • As r:=h(s) and h(s)=g(s) we see that:
                  • rV[h(s)Vg(s)V]
                • So sh1(V) and sg1(V)
                  • Notice, by continuity and V being open in (E,H) that h1(V) and g1(V) are both open in (Y,K).
                • Thus sh1(V)g1(V) (as by definition of intersection (sAB)(sAsB) - we have the RHS, so we have the left.
                  • Define W:=h1(V)g1(V), so WY
                    • notice:
                      1. that W is open in (Y,K) as by definition of a topology the intersection of open sets is open.
                      2. W (W is non-empty) as - at least - sW
                    • Define q:=p|ImV:VU be the homeomorphism of the restriction of p to V which is onto U.
                      • This means that q is injective, i.e.:
                        • v,wV[q(v)=q(w)v=w]
                      • Notice also that h(W)V and g(W)V
                      • Let wW be given
                        • By definition of being lifts: f(w)=p(h(w))=p(g(w))
                        • As h(w)V and g(w)V we see that p(h(w))=q(w) and p(g(w))=q(g(w))
                          • So f(w)=q(h(w))=q(g(w))
                            • But q is injective, so q(h(w))=q(g(w))g(w)=h(w)
                        • so we have g(w)=h(w)
                          • so wS (by definition of S)
                      • Since wW was arbitrary we have shown wW[wS]
                  • Thus WY, that is Y contains a neighbourhood of s
  • Since sS was arbitrary, we have shown sSWK[xWWY]
    • In words: that for all points in S there is a neighbourhood to that point contained entirely in S

Thus S is open, as required.

Proof that T is an open set

\newcommand{\bigudot}{ \mathchoice{\mathop{\bigcup\mkern-15mu\cdot\mkern8mu}}{\mathop{\bigcup\mkern-13mu\cdot\mkern5mu}}{\mathop{\bigcup\mkern-13mu\cdot\mkern5mu}}{\mathop{\bigcup\mkern-13mu\cdot\mkern5mu}} }\newcommand{\udot}{\cup\mkern-12.5mu\cdot\mkern6.25mu\!}\require{AMScd}\newcommand{\d}[1][]{\mathrm{d}^{#1} }

  • Let t\in T be given - Caution:The result requires that T is empty, but we can still do this[Note 2]
    • Define r_1:\eq g(t), note r_1\in E
      • Define r_2:\eq h(t), note r_2\in E
        • Define z:\eq p(r_1):\eq p(g(t))\eq f(t) by g being a lifting, note z\in X
          • Note that z:\eq p(g(t))\eq f(t)\eq p(h(t))\eq: p(r_2) by h being a lifting, so we have
            • p(r_1)\eq p(r_2)
          • As p is a covering map we see that:
            • \exists U\in\mathcal{J}[z\in U\wedge U\text{ is evenly covered by }p]
              • By evenly covered: \exists(V_\alpha)_{\alpha\in I}\subseteq\mathcal{H}\big[\bigudot_{\alpha\in I}V_\alpha\eq p^{-1}(U)\wedge \big(\forall\beta\in I[V_\beta\cong_{p\vert_{V_\beta}^\text{Im} } U]\big)\big] (note that \bigudot emphasises the union is of pairwise disjoint sets)
          • Define (V_\alpha)_{\alpha\in I} \subseteq H to be the arbitrary family of sets open in E which are the sheets of the covering of U.
            • Then:
              1. \exists\beta\in I[r_1\in V_\beta]
              2. \exists\gamma\in I[r_2\in V_\gamma]
            • Lemma: \beta\neq\gamma
              • Suppose that \beta\eq\gamma (and thus use V_\beta as the notation and discard \gamma)
                • Then r_1\in V_\beta and r_2\in V_\beta
                • Define q:\eq p\vert^\text{Im}_{V_\beta}:V_\beta\rightarrow U, so V_\beta\cong{}_q U[Note 3]
                  • Note that as it is a homeomorphism it is a bijection and as it is a bijection it is an injection, so:
                    • \forall a,b\in V_\beta[a\neq b\implies q(a)\neq q(b)][Note 4]
                  • So r_1\neq r_2\implies q(r_1)\neq q(r_2)
                    • But q(r_1)\eq p(r_1)\eq p(r_2)\eq q(r_2) (from above) - so q(r_1)\eq q(r_2)
                      • This is a contradiction, so we cannot have \beta\eq\gamma
              • Thus \beta\neq\gamma
            • By the lemma and the property of the covering sheets (that they're pairwise disjoint) we see:
              • V_\beta\cap V_\gamma\eq\emptyset (they're disjoint)
            • Define q_\beta:\eq p\vert^\text{Im}_{V_\beta}:V_\beta\rightarrow U so that V_\beta\cong_{q_\beta} U
              • Define q_\gamma:\eq p\vert^\text{Im}_{V_\gamma}:V_\gamma\rightarrow U so that V_\gamma\cong_{q_\gamma} U
                • Now V_\beta\cong_{q_\beta} U {}_{q_\gamma}\cong V_\gamma
                • Define W_\beta:\eq g^{-1}(V_\beta) - by continuity of g and the fact that V_\beta is open we see W_\beta\in\mathcal{K} , i.e. that W_\beta is open in Y
                  • Define W_\gamma:\eq h^{-1}(V_\gamma) - by continuity of h and the fact that V_\gamma is open we see W_\gamma\in\mathcal{K} , i.e. that W_\gamma is open in Y
                    • Note that t\in g^{-1}(r_1) and r_1\in V_\beta, so t\in W_\beta
                    • Note also that t\in h^{-1}(r_2) and r_2\in V_\gamma, so t\in W_\gamma
                    • Define W:\eq W_\beta\cap W_\alpha, as we're in a topology the intersection of (finitely) many open sets is again open, so we see W\in\mathcal{K} - i.e. W is open in Y.
                      • Notice that t\in W as for an intersection we see t\in W_\beta\cap W_\gamma\iff(t\in W_\beta\wedge t\in W_\gamma)
                      • Notice also that g(W)\subseteq V_\beta and h(W)\subseteq V_\gamma
                      • As V_\beta\cap V_\gamma\eq\emptyset we see h(W)\cap g(W)\eq\emptyset -
                        TODO: Make a statement page with this, I've used it before!
                        • Because h(W)\cap g(W)\eq\emptyset they cannot agree anywhere, thus W\subseteq T
                          • Since W is open and contains t we have found an open neighbourhood contained within T for each point in T
Grade: A
This page requires one or more proofs to be filled in, it is on a to-do list for being expanded with them.
Please note that this does not mean the content is unreliable. Unless there are any caveats mentioned below the statement comes from a reliable source. As always, Warnings and limitations will be clearly shown and possibly highlighted if very important (see template:Caution et al).
The message provided is:
We still need to show T is open
  • Update: almost there! Alec (talk) 15:45, 2 March 2017 (UTC)
  • Tidy up and we're done Alec (talk) 16:23, 2 March 2017 (UTC)

Notes

  1. Jump up Lee defines covering maps and spaces a little differently. He requires that for evenly covered that U be homeomorphic to each sheet, and each sheet is connected and disjoint from the others. Thus U is connected. It may not matter
    • TODO: Does it?
    But Gamelin and Green do not do this, so we can have coverings of not-connected neighbourhoods. It's worth investigating but it isn't critical to the theory.
  2. Jump up
    TODO: Investigate
  3. Jump up This is what evenly covered means, each sheet of the covering is homeomorphic to U via the restriction onto its image of the covering map
  4. Jump up As the injection page describes, this is equivalent to and is sometimes given as the following form:
    • \forall a,b\in V_\beta[q(a)\eq q(b)\implies a\eq b]
    This is just the contrapositive of what we have

References

  1. Jump up Introduction to Topology - Theodore W. Gamelin & Robert Everist Greene
  2. Jump up Introduction to Topological Manifolds - John M. Lee