src/HOL/ex/set.thy
changeset 13058 ad6106d7b4bb
parent 9100 9e081c812338
child 13107 8743cc847224
--- a/src/HOL/ex/set.thy	Thu Mar 14 16:00:29 2002 +0100
+++ b/src/HOL/ex/set.thy	Thu Mar 14 16:48:34 2002 +0100
@@ -1,4 +1,177 @@
+(*  Title:      HOL/ex/set.thy
+    ID:         $Id$
+    Author:     Tobias Nipkow and Lawrence C Paulson
+    Copyright   1991  University of Cambridge
+
+Set Theory examples: Cantor's Theorem, Schroeder-Berstein Theorem, etc.
+*)
 
 theory set = Main:
 
+text{*These two are cited in Benzmueller and Kohlhase's system description 
+of LEO, CADE-15, 1998 (pages 139-143) as theorems LEO could not prove.*}
+
+lemma "(X = Y Un Z) = (Y<=X & Z<=X & (ALL V. Y<=V & Z<=V --> X<=V))"
+by blast
+
+lemma "(X = Y Int Z) = (X<=Y & X<=Z & (ALL V. V<=Y & V<=Z --> V<=X))"
+by blast
+
+text{*trivial example of term synthesis: apparently hard for some provers!*}
+lemma "a ~= b ==> a:?X & b ~: ?X"
+by blast
+
+(** Examples for the Blast_tac paper **)
+
+text{*Union-image, called Un_Union_image on equalities.ML*}
+lemma "(UN x:C. f(x) Un g(x)) = Union(f`C)  Un  Union(g`C)"
+by blast
+
+text{*Inter-image, called Int_Inter_image on equalities.ML*}
+lemma "(INT x:C. f(x) Int g(x)) = Inter(f`C) Int Inter(g`C)"
+by blast
+
+text{*Singleton I.  Nice demonstration of blast_tac--and its limitations.
+For some unfathomable reason, UNIV_I increases the search space greatly*}
+lemma "!!S::'a set set. ALL x:S. ALL y:S. x<=y ==> EX z. S <= {z}"
+by (blast del: UNIV_I)
+
+text{*Singleton II.  variant of the benchmark above*}
+lemma "ALL x:S. Union(S) <= x ==> EX z. S <= {z}"
+by (blast del: UNIV_I)
+
+text{* A unique fixpoint theorem --- fast/best/meson all fail *}
+
+lemma "EX! x. f(g(x))=x ==> EX! y. g(f(y))=y"
+apply (erule ex1E, rule ex1I, erule arg_cong)
+apply (rule subst, assumption, erule allE, rule arg_cong, erule mp) 
+apply (erule arg_cong) 
+done
+
+
+
+text{* Cantor's Theorem: There is no surjection from a set to its powerset. *}
+
+text{*requires best-first search because it is undirectional*}
+lemma cantor1: "~ (EX f:: 'a=>'a set. ALL S. EX x. f(x) = S)"
+by best
+
+text{*This form displays the diagonal term*}
+lemma "ALL f:: 'a=>'a set. ALL x. f(x) ~= ?S(f)"
+by best
+
+text{*This form exploits the set constructs*}
+lemma "?S ~: range(f :: 'a=>'a set)"
+by (rule notI, erule rangeE, best)  
+
+text{*Or just this!*}
+lemma "?S ~: range(f :: 'a=>'a set)"
+by best
+
+text{* The Schroeder-Berstein Theorem *}
+
+lemma disj_lemma: "[| -(f`X) = g`(-X);  f(a)=g(b);  a:X |] ==> b:X"
+by blast
+
+lemma surj_if_then_else:
+     "-(f`X) = g`(-X) ==> surj(%z. if z:X then f(z) else g(z))"
+by (simp add: surj_def, blast)
+
+lemma bij_if_then_else: 
+     "[| inj_on f X;  inj_on g (-X);  -(f`X) = g`(-X);  
+         h = (%z. if z:X then f(z) else g(z)) |]        
+      ==> inj(h) & surj(h)"
+apply (unfold inj_on_def)
+apply (simp add: surj_if_then_else)
+apply (blast dest: disj_lemma sym)
+done
+
+lemma decomposition: "EX X. X = - (g`(- (f`X)))"
+apply (rule exI)
+apply (rule lfp_unfold)
+apply (rule monoI, blast) 
+done
+
+text{*Schroeder-Bernstein Theorem*}
+lemma "[| inj (f:: 'a=>'b);  inj (g:: 'b=>'a) |]  
+       ==> EX h:: 'a=>'b. inj(h) & surj(h)"
+apply (rule decomposition [THEN exE])
+apply (rule exI)
+apply (rule bij_if_then_else)
+   apply (rule_tac [4] refl)
+  apply (rule_tac [2] inj_on_inv)
+  apply (erule subset_inj_on [OF subset_UNIV]) 
+  txt{*tricky variable instantiations!*}
+ apply (erule ssubst, subst double_complement)
+ apply (rule subsetI, erule imageE, erule ssubst, rule rangeI) 
+apply (erule ssubst, subst double_complement, erule inv_image_comp [symmetric])
+done
+
+
+text{*Set variable instantiation examples from 
+W. W. Bledsoe and Guohui Feng, SET-VAR.
+JAR 11 (3), 1993, pages 293-314.
+
+Isabelle can prove the easy examples without any special mechanisms, but it
+can't prove the hard ones.
+*}
+
+text{*Example 1, page 295.*}
+lemma "(EX A. (ALL x:A. x <= (0::int)))"
+by force 
+
+text{*Example 2*}
+lemma "D : F --> (EX G. (ALL A:G. EX B:F. A <= B))";
+by force 
+
+text{*Example 3*}
+lemma "P(a) --> (EX A. (ALL x:A. P(x)) & (EX y. y:A))";
+by force 
+
+text{*Example 4*}
+lemma "a<b & b<(c::int) --> (EX A. a~:A & b:A & c~: A)"
+by force 
+
+text{*Example 5, page 298.*}
+lemma "P(f(b)) --> (EX s A. (ALL x:A. P(x)) & f(s):A)";
+by force 
+
+text{*Example 6*}
+lemma "P(f(b)) --> (EX s A. (ALL x:A. P(x)) & f(s):A)";
+by force 
+
+text{*Example 7*}
+lemma "EX A. a ~: A"
+by force 
+
+text{*Example 8*}
+lemma "(ALL u v. u < (0::int) --> u ~= abs v) --> (EX A::int set. (ALL y. abs y ~: A) & -2 : A)"
+by force  text{*not blast, which can't simplify -2<0*}
+
+text{*Example 9 omitted (requires the reals)*}
+
+text{*The paper has no Example 10!*}
+
+text{*Example 11: needs a hint*}
+lemma "(ALL A. 0:A & (ALL x:A. Suc(x):A) --> n:A) & 
+       P(0) & (ALL x. P(x) --> P(Suc(x))) --> P(n)"
+apply clarify
+apply (drule_tac x="{x. P x}" in spec)
+by force  
+
+text{*Example 12*}
+lemma "(ALL A. (0,0):A & (ALL x y. (x,y):A --> (Suc(x),Suc(y)):A) --> (n,m):A)
+       & P(n) --> P(m)"
+by auto 
+
+text{*Example EO1: typo in article, and with the obvious fix it seems
+      to require arithmetic reasoning.*}
+lemma "(ALL x. (EX u. x=2*u) = (~(EX v. Suc x = 2*v))) --> 
+       (EX A. ALL x. (x : A) = (Suc x ~: A))"
+apply clarify 
+apply (rule_tac x="{x. EX u. x = 2*u}" in exI, auto) 
+apply (case_tac v, auto)
+apply (drule_tac x="Suc v" and P="%x. ?a(x) ~= ?b(x)" in spec, force) 
+done
+
 end