Semiconjugate and commuting maps #
We define the following predicates:
Function.Semiconj
:f : α → β
semiconjugatesga : α → α
togb : β → β
iff ∘ ga = gb ∘ f
;Function.Semiconj₂
:f : α → β
semiconjugates a binary operationga : α → α → α
togb : β → β → β
iff (ga x y) = gb (f x) (f y)
;Function.Commute
:f : α → α
commutes withg : α → α
iff ∘ g = g ∘ f
, or equivalentlySemiconj f g g
.
We say that f : α → β
semiconjugates ga : α → α
to gb : β → β
if f ∘ ga = gb ∘ f
.
We use ∀ x, f (ga x) = gb (f x)
as the definition, so given h : Function.Semiconj f ga gb
and
a : α
, we have h a : f (ga a) = gb (f a)
and h.comp_eq : f ∘ ga = gb ∘ f
.
Equations
- Function.Semiconj f ga gb = ∀ (x : α), f (ga x) = gb (f x)
Instances For
Two maps f g : α → α
commute if f (g x) = g (f x)
for all x : α
.
Given h : Function.commute f g
and a : α
, we have h a : f (g a) = g (f a)
and
h.comp_eq : f ∘ g = g ∘ f
.
Equations
- Function.Commute f g = Function.Semiconj f g g
Instances For
Reinterpret Function.Semiconj f g g
as Function.Commute f g
. These two predicates are
definitionally equal but have different dot-notation lemmas.
Reinterpret Function.Commute f g
as Function.Semiconj f g g
. These two predicates are
definitionally equal but have different dot-notation lemmas.
A map f
semiconjugates a binary operation ga
to a binary operation gb
if
for all x
, y
we have f (ga x y) = gb (f x) (f y)
. E.g., a MonoidHom
semiconjugates (*)
to (*)
.
Equations
- Function.Semiconj₂ f ga gb = ∀ (x y : α), f (ga x y) = gb (f x) (f y)