Page 134 - DMTH503_TOPOLOGY
P. 134

Topology




                    Notes          so that
                                                               d(x, A) – d(y, A) d(x, y).
                                   The same inequality holds with x and y interchanged, continuity of the function d(x, A) follows.

                                   Now we introduce the notion of Lebesgue number. Recall that the diameter of a bounded subset
                                   A of a metric space (X, d) is the number
                                                               sup {d(a , a ) | a , a  A}
                                                                     1  2   1  2
                                   Lemma (1) (The Lebesgue number Lemma): Let  be an open covering of the metric space (X, d). If
                                   X is compact, there is a 0 such that for each subset of X having diameter less than , there exists
                                   an element of  containing it.
                                   The number  is called a Lebesgue number for the covering .

                                   Proof: Let  be an open covering of X. If X itself is an element of , then any positive number is
                                   a Lebesgue number of . So assume X is not an element of .
                                   Choose a finite subcollection {A , ..., A } of  that covers X. For each i, set C  = X – A , and define
                                                            1    n                            i      i
                                   f: X   be letting f(x) be the average of the numbers d(x, C ). That is,
                                                                                   i
                                                                       1  n
                                                                  f(x) =    (x, c )
                                                                       n   i 1  i
                                   We  show that  f(x) >  0  for  all x.  Given  x    X,  choose  i so  that  x    A .  Then choose    so
                                                                                              i
                                   -neighborhood of x lies in A . Then d(x, c ) , so that f(x) /n.
                                                          i          i
                                   Since f  is continuous,  it has a minimum value    we  show that    is our required  Lebesgue
                                                                           i
                                   number. Let B be a subset of X of diameter less that . Choose a point x  of B; then B lies in the
                                                                                            0
                                   -neighborhood of x . Now
                                                   0
                                                                  f(x )  d(x , C ),
                                                                      0     0  m
                                   where  d(x , C ) is the largest of the number d(x , C ). Then the -neighborhood of x  is contained
                                           0  m                         0  i                        0
                                   in the element A  – X – C  of one covering .
                                                m      m
                                   Definition: Uniformly Continuous
                                   A function F from the metric space (X, d ) to the metric (Y, d ) is said to be uniformly continuous
                                                                  X               Y
                                   if given  > 0, there is a  > 0 such that for every pair of points x , x  of X,
                                                                                      0  1
                                                           d (x , x ) < d   d (f(x ), f(x )) < .
                                                            x  0  1      Y   0   1
                                   Theorem 7: Uniform Continuity Theorem

                                   Let f: X Y be a continuous map of the compact metric space (X, d ) to be metric space (Y, d ).
                                                                                         x                   y
                                   Then f is uniformly continuous.
                                   Proof: Given > 0, take the open covering of Y by balls B (y, /2) of radius /2. Let A be the
                                   open covering of X by the inverse images of these balls under  f. Choose  to be a Lebesgue
                                   number for the covering A. Then if x  and x  are two points of X such that dx(x , x ) < , the two
                                                                1    2                            1  2
                                   point set {x , x } has diameter less than . So that its image {f(x ), f(x )} lies in some ball B (y, /2).
                                           1  2                                    1   2
                                   Then dy (f(x ), f(x ) < , as desired.
                                            1    2
                                   Finally, we prove that the real numbers are uncountable. The interesting thing about this proof
                                   is that it involves no algebra at all-no  decimal or binary expansions of real numbers or the
                                   like-just the other properties of  .







          128                               LOVELY PROFESSIONAL UNIVERSITY
   129   130   131   132   133   134   135   136   137   138   139