Page 263 - DMTH503_TOPOLOGY
P. 263

Unit 32: Introduction to Dimension Theory




          Consider a probability vector  p = (p , …, p ) and define the Bernoulli measure of any level  Notes
                                        0     n–1
          cylinder to be,

                            [ ( i ,  ,i k 1 ])  = p k 0 0 ( i ) p k 1 1 ( i )  p k - n 1 ( i ) .
                                                  n 1
                             0
                                  -
                                                   -
          A probability measure  on  is said to be invariant under the shift map if for any Borel set
                                                                         –1
                        –1
          B  X, (B) =( (B)). We say thatis ergodic if any Borel set Bsuch that  (X) = X satisfies
          (X) = 0 or (X) = 1. A Bernoulli measure is both invariant and ergodic.
          32.2 Summary
              Criteria for defining a dimension
               (i)  When X is a manifold then the value of the dimension is an integer which coincides
                    with the usual notion of dimension;

               (ii)  For more general sets X we can have “fractional” dimensional; and
               (iii)  Points and countable unions of points, have zero dimension.
              For a given probability measure , we define the Hausdorff dimension of the measure by

                       dim () = inf {dim (X) : (X) = 1}.
                          H           H
              The upper and lower pointwise dimensions of a measure  are measurable functions,  d ,
                                                                                     
               d : X{¥} defined by
                
                                       log (B(x,r))
                                          
                         d (x)  = lim  sup        and
                          
                                r  0     log r
                                      log (B(x,r))
                                         
                         d (x)  = lim inf
                          
                                r  0    log r
          32.3 Keywords

          Countable Set: A set is countable if it is non-empty and finite or if it is countably infinite.
          Hausdorff Space: A topological space (X, T) is called Hausdorff space if given a pair of distinct
          points x, yX,
                               G, HT  s.t.  xG, yH, GÇH = .
          Iterated Function Scheme: An iterated function scheme on an open setR  consists of a family
                                                                      d
          of contractions T , …, T  :  .
                        1    k
          Open Set: Any set AT is called an open set.
          Subcover: Let (X, T) be a topological space and AX. Let G denote a family of subsets of X. If
          G G s.t. G  is a finite set and that {G : GG } is a cover of A then G  is called a finite subcover
            1       1                          1                 1
          of the original cover.
          32.4 Review Questions


          1.   Write a short note on Dimension Theory.
          2.   State Besicovitch covering lemma.




                                           LOVELY PROFESSIONAL UNIVERSITY                                   257
   258   259   260   261   262   263   264   265