The Mizar article:

Fix Point Theorem for Compact Spaces

by
Alicia de la Cruz

Received July 17, 1991

Copyright (c) 1991 Association of Mizar Users

MML identifier: ALI2
[ MML identifier index ]


environ

 vocabulary METRIC_1, FINSET_1, BOOLE, ZFMISC_1, FUNCT_1, PRE_TOPC, FUNCOP_1,
      RELAT_1, ARYTM_3, PCOMPS_1, COMPTS_1, SETFAM_1, POWER, SUBSET_1, ARYTM_1,
      SEQ_1, SEQ_2, ORDINAL2, SEQM_3, ALI2, HAHNBAN;
 notation TARSKI, XBOOLE_0, SUBSET_1, ORDINAL1, XREAL_0, FINSET_1, SETFAM_1,
      METRIC_1, RELAT_1, FUNCT_2, STRUCT_0, PRE_TOPC, POWER, FUNCOP_1,
      COMPTS_1, PCOMPS_1, TOPS_2, SEQ_1, SEQ_2, SEQM_3, REAL_1, NAT_1;
 constructors FINSET_1, POWER, FUNCOP_1, COMPTS_1, PCOMPS_1, TOPS_2, SEQ_2,
      SEQM_3, REAL_1, NAT_1, PARTFUN1, MEMBERED;
 clusters STRUCT_0, PCOMPS_1, FUNCOP_1, XREAL_0, SEQ_1, METRIC_1, MEMBERED,
      ZFMISC_1;
 requirements REAL, NUMERALS, SUBSET, BOOLE, ARITHM;
 definitions COMPTS_1, TARSKI, TOPS_2, ORDINAL1, XBOOLE_0;
 theorems METRIC_1, SUBSET_1, REAL_1, PCOMPS_1, REAL_2, COMPTS_1, POWER, SEQ_2,
      SEQ_4, SERIES_1, SEQM_3, AXIOMS, SETFAM_1, TARSKI, SEQ_1, PRE_TOPC,
      TOPS_1, SQUARE_1, FUNCOP_1, FUNCT_2, RELSET_1, ORDINAL1, XBOOLE_0,
      XBOOLE_1, XCMPLX_1;
 schemes SETFAM_1, SEQ_1, DOMAIN_1, FINSET_1, NAT_1;

begin
  reserve M for non empty MetrSpace;

theorem Th1:
for F being set st F is finite & F <> {} & F is c=-linear
 ex m being set st m in F & for C being set st C in F holds m c= C
 proof
  defpred P[set] means $1 <> {} implies
   ex m being set st m in $1 & for C being set st C in $1 holds m c= C;
  let F be set such that
  A1: F is finite and
  A2: F <> {} and
  A3: F is c=-linear;
  A4: P[{}];
  A5: now let x,B be set such that A6:x in F & B c= F & P[B];
      now per cases;                         :: we have to prove P[BU{x}]
     suppose A7:not ex y being set st y in B & y c=x;
      assume B \/ {x} <> {};
      take m = x;
         x in {x} by TARSKI:def 1;
      hence m in B \/ {x} by XBOOLE_0:def 2;
      let C be set;
      assume C in B \/ {x};
      then A8: C in B or C in {x} by XBOOLE_0:def 2;
      then C,x are_c=-comparable by A3,A6,ORDINAL1:def 9,TARSKI:def 1;
      then C c= x or x c= C by XBOOLE_0:def 9;
       hence m c= C by A7,A8,TARSKI:def 1;
     suppose ex y being set st y in B & y c=x;
     then consider y being set such that A9: y in B & y c=x;
      assume B \/ {x} <> {};
       consider m being set such that A10:m in B and
       A11: for C being set st C in B holds m c= C by A6,A9;
         m c= y by A9,A11;
       then A12:m c= x by A9,XBOOLE_1:1;
        take m;
      thus m in B \/ {x} by A10,XBOOLE_0:def 2;
      let C be set;
      assume C in B \/ {x};
      then C in B or C in {x} by XBOOLE_0:def 2;
      hence m c= C by A11,A12,TARSKI:def 1;
    end;
   hence P[B \/ {x}];
   end;
     P[F] from Finite(A1,A4,A5);
  hence thesis by A2;
 end;

definition let M be non empty MetrSpace;
  mode contraction of M -> Function of the carrier of M, the carrier of M
    means :Def1:
  ex L being Real st 0 < L & L < 1 &
    for x,y being Point of M holds dist(it.x,it.y) <= L*dist(x,y);
existence
proof
 consider x being Point of M;
   (the carrier of M) --> x is Function of the carrier of M, the carrier of M
 proof
A1: dom ((the carrier of M) --> x) = the carrier of M by FUNCOP_1:19;
 rng ((the carrier of M) --> x) c= {x} by FUNCOP_1:19;
 then rng ((the carrier of M) --> x) c= the carrier of M by XBOOLE_1:1;
  hence thesis by A1,FUNCT_2:def 1,RELSET_1:11;
 end;
 then reconsider f = (the carrier of M) --> x as Function of the carrier of M,
   the carrier of M;
 take f, 1/2;
 thus 0<1/2 & 1/2<1;
 let z,y be Point of M;
   f.z=x & f.y=x by FUNCOP_1:13;
 then A2: dist(f.z,f.y) = 0 by METRIC_1:1;
   dist(z,y)>=0 by METRIC_1:5;
 hence dist(f.z,f.y)<=(1/2)*dist(z,y) by A2,REAL_2:121;
end;
end;

theorem
  for f being contraction of M st TopSpaceMetr(M) is compact
ex c being Point of M st f.c =c &            :: exists a fix point
for x being Point of M st f.x=x holds x=c    :: exactly 1 fix point
proof
  let f be contraction of M;
  consider L being Real such that
  A1: 0<L & L<1 and
  A2: for x,y being Point of M holds dist(f.x,f.y)<=L*dist(x,y) by Def1;
  assume A3: TopSpaceMetr(M) is compact;
  consider x0 being Point of M;
  set a=dist(x0,f.x0);
    now assume a <> 0;
   defpred _P[set] means ex n being Nat st $1 = { x where x is Point of M :
   dist(x,f.x) <= a*L to_power n};
  consider F being Subset-Family of TopSpaceMetr(M)
  such that A4: for B being Subset of TopSpaceMetr(M) holds
  B in F iff _P[B] from SubFamEx;
A5:   TopSpaceMetr(M)=TopStruct (#the carrier of M,Family_open_set(M)#)
      by PCOMPS_1:def 6;
    defpred _P[Point of M] means dist($1,f.($1)) <= a*L to_power (0+1);
A6: {x where x is Point of M:_P[x]}is Subset of M from SubsetD;
  A7: F is centered
  proof
   thus F <> {} by A4,A5,A6;
   let G be set
   such that A8: G <> {}
   and A9: G c= F
   and A10: G is finite;
     G is c=-linear
   proof
   let B,C be set; assume B in G & C in G;
   then A11:B in F & C in F by A9;
   then consider n being Nat such that A12: B= { x where x is Point of M :
   dist(x,f.x) <= a*L to_power n} by A4;
   consider m being Nat such that A13: C= { x where x is Point of M :
   dist(x,f.x) <= a*L to_power m} by A4,A11;
   A14: for n,m being Nat st n<=m holds L to_power m <= L to_power n
   proof
    let n,m be Nat such that A15:n<=m;
      now per cases by A15,REAL_1:def 5;
    suppose n<m;
     hence L to_power n >= L to_power m by A1,POWER:45;
    suppose n=m;
     hence L to_power n >= L to_power m;
    end;
    hence thesis;
   end;

   A16: for n,m being Nat st n<=m holds a*L to_power m <= a*L to_power n
   proof
   let n,m be Nat such that A17:n<=m;
    A18:a>=0 by METRIC_1:5;
      L to_power m <= L to_power n by A14,A17;
    hence a*L to_power m <= a*L to_power n by A18,AXIOMS:25;
   end;

     now per cases;
   case A19:n<=m;
   thus C c= B
    proof
     let y be set;
      assume y in C;
      then consider x being Point of M such that
A20:    y = x and
A21:    dist(x,f.x) <= a*L to_power m by A13;
         a*L to_power m <= a*L to_power n by A16,A19;
       then dist(x,f.x) <= a*L to_power n by A21,AXIOMS:22;
      hence y in B by A12,A20;
    end;
   case A22:m<=n;
    thus B c= C
    proof
     let y be set;
      assume y in B;
      then consider x being Point of M such that
A23:    y = x and
A24:    dist(x,f.x) <= a*L to_power n by A12;
         a*L to_power n <= a*L to_power m by A16,A22;
       then dist(x,f.x) <= a*L to_power m by A24,AXIOMS:22;
      hence y in C by A13,A23;
    end;
   end;
   hence B c= C or C c= B;
   end;
   then consider m being set such that
   A25: m in G and
   A26: for C being set st C in G holds m c= C by A8,A10,Th1;
   A27: m c= meet G by A8,A26,SETFAM_1:6;
   A28: m in F by A9,A25;
   defpred _P[Nat] means
     { x where x is Point of M : dist(x,f.x) <= a*L to_power $1}<>{};
     dist(x0,f.x0) = a*1
                .= a*L to_power 0 by POWER:29;
   then x0 in { x where x is Point of M : dist(x,f.x) <= a*L to_power 0};
   then A29: _P[0];
   A30:for k being Nat st _P[k] holds _P[k+1]
   proof
   let k be Nat;
    consider z being Element of
     { x where x is Point of M : dist(x,f.x) <= a*L to_power k};
   assume
     { x where x is Point of M : dist(x,f.x) <= a*L to_power k}<>{};
   then z in { x where x is Point of M : dist(x,f.x) <= a*L to_power k};
   then consider y being Point of M such that
     y=z and A31:dist(y,f.y)<= a*L to_power k;
   A32:L*dist(y,f.y) <= L*(a*L to_power k) by A1,A31,AXIOMS:25;
   A33:L*(a*L to_power k)=a*(L*L to_power k) by XCMPLX_1:4
                     .=a*(L to_power k*L to_power 1) by POWER:30
                     .= a*L to_power (k+1) by A1,POWER:32;
     dist(f.y,f.(f.y)) <= L*dist(y,f.y) by A2;
   then dist(f.y,f.(f.y)) <= a*L to_power (k+1) by A32,A33,AXIOMS:22;
   then f.y in { x where x is Point of M : dist(x,f.x) <= a*L to_power (k+1)};
   hence { x where x is Point of M : dist(x,f.x) <= a*L to_power (k+1)}<>{};
   end;
     for k being Nat holds _P[k] from Ind(A29,A30);
   then m <> {} by A4,A28;
   hence meet G <> {} by A27,XBOOLE_1:3;
  end;
    F is closed
   proof
   let B be Subset of TopSpaceMetr(M);
   assume B in F;
   then consider n being Nat such that
   A34:B= { x where x is Point of M : dist(x,f.x) <= a*L to_power n} by A4;
   A35: TopSpaceMetr(M)=TopStruct (#the carrier of M,Family_open_set(M)#)
     by PCOMPS_1:def 6;
   then reconsider V = B` as Subset of M;

     for x being Point of M st x in V ex r being Real st r>0 & Ball(x,r) c= V
   proof
   let x be Point of M such that A36:x in V;
   take r = (dist(x,f.x)-a*L to_power n)/2;
     2*r = dist(x,f.x)-a*L to_power n by XCMPLX_1:88;
   then A37: dist(x,f.x)-2*r = a*L to_power n by XCMPLX_1:18;
     not x in B by A36,SUBSET_1:54;
   then dist(x,f.x)>a*L to_power n by A34;
   then dist(x,f.x)-a*L to_power n>0 by SQUARE_1:11;
   hence r>0 by SEQ_2:3;
   let z be set;
   assume A38:z in Ball(x,r);
   then reconsider y=z as Point of M;
   A39: dist(x,y)<r by A38,METRIC_1:12;
     dist(x,y) + dist(y,f.y) >= dist(x,f.y) by METRIC_1:4;
   then A40:  (dist(x,y) + dist(y,f.y)) + dist(f.y,f.x) >=
                       dist(x,f.y) + dist(f.y,f.x) by AXIOMS:24;
     dist(x,f.y) + dist(f.y,f.x) >= dist(x,f.x) by METRIC_1:4;
   then dist(y,f.y)+dist(x,y)+dist(f.y,f.x)>=dist(x,f.x) by A40,AXIOMS:22;
   then A41: dist(y,f.y)+(dist(x,y)+dist(f.y,f.x))>=dist(x,f.x) by XCMPLX_1:1;
   A42: dist(f.y,f.x)<=L*dist(y,x) by A2;
     dist(y,x)>=0 by METRIC_1:5;
   then L*dist(y,x)<=dist(y,x) by A1,REAL_2:147;
   then dist(f.y,f.x)<=dist(y,x) by A42,AXIOMS:22;
   then A43: dist(f.y,f.x)+dist(y,x) <= dist(y,x)+dist(y,x) by AXIOMS:24;
     2*dist(x,y)<2*r by A39,REAL_1:70;
   then A44: dist(y,f.y) + 2*dist(x,y)< dist(y,f.y) + 2*r by REAL_1:53;
     dist(y,x) + dist(f.y,f.x) <= 2*dist(y,x) by A43,XCMPLX_1:11;
   then dist(y,f.y) + (dist(y,x) + dist(f.y,f.x)) <=
                      dist(y,f.y) + 2*dist(y,x) by REAL_1:55;
   then dist(y,f.y)+2*dist(x,y)>=dist(x,f.x) by A41,AXIOMS:22;
   then dist(y,f.y)+2*r>dist(x,f.x) by A44,AXIOMS:22;
   then not ex x being Point of M st y = x & dist(x,f.x)<= a*L to_power n by
A37,REAL_1:84;
   then not y in B by A34;
   hence z in V by A35,SUBSET_1:50;
   end;
   then B` in Family_open_set(M) by PCOMPS_1:def 5;
   then B` in Family_open_set(M);
   then B` is open by A35,PRE_TOPC:def 5;
   hence B is closed by TOPS_1:29;
   end;
  then meet F <> {} by A3,A7,COMPTS_1:13;
  then consider c' being Point of TopSpaceMetr(M)
     such that A45: c' in meet F by SUBSET_1:10;
  reconsider c = c' as Point of M by A5;
  deffunc _F(Nat) = L to_power ($1+1);
  consider s' being Real_Sequence such that
  A46:for n being Nat holds s'.n= _F(n) from ExRealSeq;
  set s = a (#) s';
  A47: s' is convergent & lim s'=0 by A1,A46,SERIES_1:1;
  then A48: s is convergent by SEQ_2:21;
  A49: lim s = a*0 by A47,SEQ_2:22 .= 0;
  deffunc _F(set) = dist(c,f.c);
  consider r being Real_Sequence such that
  A50: for n being Nat holds r.n=_F(n) from ExRealSeq;
  A51: r is constant by A50,SEQM_3:def 6;
  then A52: r is convergent by SEQ_4:39;
    now let n be Nat;
     defpred _P[Point of M] means dist($1,f.$1) <= a*L to_power (n+1);
    set B = { x where x is Point of M : _P[x]};
      B is Subset of M from SubsetD;
    then B in F by A4,A5;
    then c in B by A45,SETFAM_1:def 1;
    then A53: ex x being Point of M st c = x & dist(x,f.x) <= a*L to_power (n+1
);
       s.n = a*s'.n by SEQ_1:13
      .= a*L to_power (n+1) by A46;
   hence r.n <= s.n by A50,A53;
  end;
  then A54: lim r <= lim s by A48,A52,SEQ_2:32;
    r.0=dist(c,f.c) by A50;
  then dist(c,f.c)<=0 & dist(c,f.c)>=0 by A49,A51,A54,METRIC_1:5,SEQ_4:40;
  then dist(c,f.c)=0;
  hence ex c being Point of M st dist(c,f.c) = 0;
  end;
  then consider c being Point of M such that
  A55: dist(c,f.c) = 0;
  take c;
  thus A56: f.c =c by A55,METRIC_1:2;
  let x be Point of M; assume A57: f.x=x;
  assume x<>c;
  then A58: dist(x,c)<>0 by METRIC_1:2;
    dist(x,c)>=0 by METRIC_1:5;
  then L*dist(x,c)<dist(x,c) by A1,A58,REAL_2:145;
  hence contradiction by A2,A56,A57;
end;

Back to top