Nuprl Lemma : unique-limit
∀[x:ℕ ⟶ ℝ]. ∀[y1,y2:ℝ].  (y1 = y2) supposing (lim n→∞.x[n] = y2 and lim n→∞.x[n] = y1)
Proof
Definitions occuring in Statement : 
converges-to: lim n→∞.x[n] = y
, 
req: x = y
, 
real: ℝ
, 
nat: ℕ
, 
uimplies: b supposing a
, 
uall: ∀[x:A]. B[x]
, 
so_apply: x[s]
, 
function: x:A ⟶ B[x]
Definitions unfolded in proof : 
uall: ∀[x:A]. B[x]
, 
member: t ∈ T
, 
uimplies: b supposing a
, 
converges-to: lim n→∞.x[n] = y
, 
uiff: uiff(P;Q)
, 
and: P ∧ Q
, 
all: ∀x:A. B[x]
, 
nat_plus: ℕ+
, 
less_than: a < b
, 
squash: ↓T
, 
less_than': less_than'(a;b)
, 
true: True
, 
prop: ℙ
, 
sq_exists: ∃x:{A| B[x]}
, 
so_apply: x[s]
, 
nat: ℕ
, 
implies: P 
⇒ Q
, 
guard: {T}
, 
ge: i ≥ j 
, 
decidable: Dec(P)
, 
or: P ∨ Q
, 
satisfiable_int_formula: satisfiable_int_formula(fmla)
, 
exists: ∃x:A. B[x]
, 
false: False
, 
not: ¬A
, 
top: Top
, 
rneq: x ≠ y
, 
iff: P 
⇐⇒ Q
, 
rev_implies: P 
⇐ Q
, 
sq_stable: SqStable(P)
, 
rleq: x ≤ y
, 
rnonneg: rnonneg(x)
, 
le: A ≤ B
, 
subtype_rel: A ⊆r B
, 
so_lambda: λ2x.t[x]
, 
rev_uimplies: rev_uimplies(P;Q)
, 
rge: x ≥ y
, 
rsub: x - y
Lemmas referenced : 
infinitesmal-difference, 
mul_nat_plus, 
less_than_wf, 
sq_stable__rleq, 
rabs_wf, 
rsub_wf, 
nat_wf, 
imax_wf, 
imax_nat, 
nat_properties, 
nat_plus_properties, 
decidable__le, 
satisfiable-full-omega-tt, 
intformand_wf, 
intformnot_wf, 
intformle_wf, 
itermConstant_wf, 
itermVar_wf, 
intformeq_wf, 
int_formula_prop_and_lemma, 
int_formula_prop_not_lemma, 
int_formula_prop_le_lemma, 
int_term_value_constant_lemma, 
int_term_value_var_lemma, 
int_formula_prop_eq_lemma, 
int_formula_prop_wf, 
equal_wf, 
le_wf, 
rdiv_wf, 
int-to-real_wf, 
rless-int, 
decidable__lt, 
intformless_wf, 
itermMultiply_wf, 
int_formula_prop_less_lemma, 
int_term_value_mul_lemma, 
rless_wf, 
imax_ub, 
less_than'_wf, 
nat_plus_wf, 
req_witness, 
converges-to_wf, 
real_wf, 
radd_wf, 
rleq_wf, 
rminus_wf, 
rleq_weakening_equal, 
rleq_functionality_wrt_implies, 
r-triangle-inequality, 
uiff_transitivity, 
rleq_functionality, 
req_weakening, 
rabs_functionality, 
req_inversion, 
radd-assoc, 
req_transitivity, 
radd-ac, 
radd_functionality, 
radd-rminus-assoc, 
radd_functionality_wrt_rleq, 
rabs-difference-symmetry, 
rmul_wf, 
rleq-int-fractions, 
rmul-identity1, 
rmul-distrib2, 
rmul_functionality, 
radd-int, 
rmul-int-rdiv
Rules used in proof : 
sqequalSubstitution, 
sqequalTransitivity, 
computationStep, 
sqequalReflexivity, 
isect_memberFormation, 
introduction, 
cut, 
sqequalHypSubstitution, 
extract_by_obid, 
isectElimination, 
thin, 
hypothesisEquality, 
productElimination, 
independent_isectElimination, 
hypothesis, 
dependent_functionElimination, 
dependent_set_memberEquality, 
natural_numberEquality, 
sqequalRule, 
independent_pairFormation, 
imageMemberEquality, 
baseClosed, 
because_Cache, 
setElimination, 
rename, 
applyEquality, 
functionExtensionality, 
lambdaFormation, 
equalityTransitivity, 
equalitySymmetry, 
applyLambdaEquality, 
unionElimination, 
dependent_pairFormation, 
lambdaEquality, 
int_eqEquality, 
intEquality, 
isect_memberEquality, 
voidElimination, 
voidEquality, 
computeAll, 
independent_functionElimination, 
multiplyEquality, 
inrFormation, 
imageElimination, 
inlFormation, 
independent_pairEquality, 
minusEquality, 
axiomEquality, 
functionEquality, 
addEquality
Latex:
\mforall{}[x:\mBbbN{}  {}\mrightarrow{}  \mBbbR{}].  \mforall{}[y1,y2:\mBbbR{}].    (y1  =  y2)  supposing  (lim  n\mrightarrow{}\minfty{}.x[n]  =  y2  and  lim  n\mrightarrow{}\minfty{}.x[n]  =  y1)
Date html generated:
2017_10_03-AM-08_52_28
Last ObjectModification:
2017_07_28-AM-07_35_11
Theory : reals
Home
Index