4

In chapter 2.2 of

Giacobazzi, Roberto; Ranzato, Francesco, Uniform closures: Order-theoretically reconstructing logic program semantics and abstract domain refinements, Inf. Comput. 145, No.2, 153-190 (1998). ZBL0921.68057.

it's said:

An (upper) closure operator (or simply closure) on a poset $C$ is an operator $\rho:C \to C$ monotone, idempotent and extensive (i.e., $\forall x \in C . x \le \rho(x)$). We denote by $uco(C)$ the set of all closure operators on the poset $C$. If $C$ is a complete lattice then each closure operator $uco(C)$ is uniquely determined by the set of its fixpoints, which is its image $\rho(C)$

Where can I find a proof of the phrase in bold?

incud
  • 551
  • 7
  • 18

1 Answers1

4

There is an intimate connection between closure operators and complete lattices.

Given a closure operator $C$ on a set $X$, we can construct a complete lattice $L(C)$ as follows:

  • The points of the lattice are the fixpoints of $C$.
  • The meet of the lattice is $A \wedge B = A \cap B$.
  • The join of the lattice is $A \vee B = C(A \cup B)$.

Conversely, given a complete lattice $L$ of subsets of $X$, we can construct a closure operator $C:=C(L)$ on $X$ using the formula $$ C(A) = \bigcap \{ B \in L : B \supseteq A \}. $$

It turns out that these two operators are inverses: $C(L(C)) = C$ and $L(C(L)) = L$.

In particular, this shows that $C$ is determined by the set $L(C)$ of fixpoints, since $$ C(A) = \bigcap \{B \in L(C) : B \supseteq A\}. $$

(This information is taken from the sample chapter of M-solid varieties of algebras by Koppitz and Denecke.)

Yuval Filmus
  • 280,205
  • 27
  • 317
  • 514