Nuprl Lemma : nc-e-comp-nc-1
∀[I:fset(ℕ)]. ∀[i,j:{j:ℕ| ¬j ∈ I} ].  ((i1) = e(i;j) ⋅ (j1) ∈ I ⟶ I+i)
Proof
Definitions occuring in Statement : 
nc-e: e(i;j)
, 
nc-1: (i1)
, 
add-name: I+i
, 
nh-comp: g ⋅ f
, 
names-hom: I ⟶ J
, 
fset-member: a ∈ s
, 
fset: fset(T)
, 
int-deq: IntDeq
, 
nat: ℕ
, 
uall: ∀[x:A]. B[x]
, 
not: ¬A
, 
set: {x:A| B[x]} 
, 
equal: s = t ∈ T
Definitions unfolded in proof : 
uall: ∀[x:A]. B[x]
, 
member: t ∈ T
, 
subtype_rel: A ⊆r B
, 
prop: ℙ
, 
uimplies: b supposing a
, 
nat: ℕ
, 
so_lambda: λ2x.t[x]
, 
so_apply: x[s]
Lemmas referenced : 
nc-1-as-nc-p, 
fset_wf, 
strong-subtype-self, 
le_wf, 
strong-subtype-set3, 
strong-subtype-deq-subtype, 
int-deq_wf, 
fset-member_wf, 
not_wf, 
nat_wf, 
dM1_wf, 
nc-e-comp-nc-p
Rules used in proof : 
cut, 
lemma_by_obid, 
sqequalSubstitution, 
sqequalTransitivity, 
computationStep, 
sqequalReflexivity, 
isect_memberFormation, 
hypothesis, 
sqequalHypSubstitution, 
isectElimination, 
thin, 
hypothesisEquality, 
setEquality, 
applyEquality, 
because_Cache, 
sqequalRule, 
intEquality, 
independent_isectElimination, 
lambdaEquality, 
natural_numberEquality, 
setElimination, 
rename, 
equalitySymmetry
Latex:
\mforall{}[I:fset(\mBbbN{})].  \mforall{}[i,j:\{j:\mBbbN{}|  \mneg{}j  \mmember{}  I\}  ].    ((i1)  =  e(i;j)  \mcdot{}  (j1))
Date html generated:
2016_05_18-PM-00_03_29
Last ObjectModification:
2016_02_05-PM-02_15_28
Theory : cubical!type!theory
Home
Index