TPTP Problem File: SEU168+1.p
View Solutions
- Solve Problem
%------------------------------------------------------------------------------
% File : SEU168+1 : TPTP v9.0.0. Released v3.3.0.
% Domain : Set theory
% Problem : MPTP bushy problem t136_zfmisc_1
% Version : [Urb07] axioms : Especial.
% English :
% Refs : [Ban01] Bancerek et al. (2001), On the Characterizations of Co
% : [Urb07] Urban (2006), Email to G. Sutcliffe
% Source : [Urb07]
% Names : bushy-t136_zfmisc_1 [Urb07]
% Status : Theorem
% Rating : 0.55 v9.0.0, 0.56 v8.1.0, 0.58 v7.5.0, 0.56 v7.4.0, 0.53 v7.3.0, 0.52 v7.2.0, 0.48 v7.0.0, 0.50 v6.3.0, 0.46 v6.2.0, 0.60 v6.1.0, 0.67 v6.0.0, 0.70 v5.5.0, 0.67 v5.4.0, 0.68 v5.3.0, 0.70 v5.2.0, 0.55 v5.1.0, 0.57 v5.0.0, 0.58 v4.1.0, 0.57 v4.0.0, 0.58 v3.7.0, 0.65 v3.5.0, 0.68 v3.4.0, 0.74 v3.3.0
% Syntax : Number of formulae : 7 ( 2 unt; 0 def)
% Number of atoms : 30 ( 1 equ)
% Maximal formula atoms : 11 ( 4 avg)
% Number of connectives : 32 ( 9 ~; 0 |; 14 &)
% ( 3 <=>; 6 =>; 0 <=; 0 <~>)
% Maximal formula depth : 14 ( 7 avg)
% Maximal term depth : 2 ( 1 avg)
% Number of predicates : 5 ( 3 usr; 1 prp; 0-2 aty)
% Number of functors : 1 ( 1 usr; 0 con; 1-1 aty)
% Number of variables : 24 ( 22 !; 2 ?)
% SPC : FOF_THM_RFO_SEQ
% Comments : Translated by MPTP 0.2 from the original problem in the Mizar
% library, www.mizar.org
%------------------------------------------------------------------------------
fof(antisymmetry_r2_hidden,axiom,
! [A,B] :
( in(A,B)
=> ~ in(B,A) ) ).
fof(d1_zfmisc_1,axiom,
! [A,B] :
( B = powerset(A)
<=> ! [C] :
( in(C,B)
<=> subset(C,A) ) ) ).
fof(d3_tarski,axiom,
! [A,B] :
( subset(A,B)
<=> ! [C] :
( in(C,A)
=> in(C,B) ) ) ).
fof(dt_k1_zfmisc_1,axiom,
$true ).
fof(reflexivity_r1_tarski,axiom,
! [A,B] : subset(A,A) ).
fof(t136_zfmisc_1,conjecture,
! [A] :
? [B] :
( in(A,B)
& ! [C,D] :
( ( in(C,B)
& subset(D,C) )
=> in(D,B) )
& ! [C] :
( in(C,B)
=> in(powerset(C),B) )
& ! [C] :
~ ( subset(C,B)
& ~ are_equipotent(C,B)
& ~ in(C,B) ) ) ).
fof(t9_tarski,axiom,
! [A] :
? [B] :
( in(A,B)
& ! [C,D] :
( ( in(C,B)
& subset(D,C) )
=> in(D,B) )
& ! [C] :
~ ( in(C,B)
& ! [D] :
~ ( in(D,B)
& ! [E] :
( subset(E,C)
=> in(E,D) ) ) )
& ! [C] :
~ ( subset(C,B)
& ~ are_equipotent(C,B)
& ~ in(C,B) ) ) ).
%------------------------------------------------------------------------------