AFTD Auto-Formalizing Theoretical Domains

is_cohen_macaulay_two_complex

def

A finite simplicial complex given by a set of facets in a linearly ordered vertex type V is defined to be Cohen-Macaulay over a field k if: (1) facets is nonempty; (2) every facet has cardinality 3; (3) the 1-skeleton graph is connected on its vertices; (4) for every vertex v in the complex and all vertices u and w with u ≠ v and w ≠ v such that {u, v} and {w, v} are each a subset of some facet in facets, u and w are reachable from each other in the link graph of v (the simple graph with edges {a, b} such that {a, b, v} is in facets); and (5) every 1-cycle over k supported on ordered edges u < v is the boundary of some 2-chain over k supported on ordered triangles u < v < w.

Statement

def is_cohen_macaulay_two_complex {V : Type*} [DecidableEq V] [LinearOrder V]
    (facets : Finset (Finset V)) (k : Type*) [Field k] : Prop
Defined in
arXiv:2610.27083 Question 6.2; Duval et al. (2016) Adv. Math. 299, 381-395
Verified
24 Oct 2026
Axioms
none listed
Built on 2
  • TotalPreorderA total preorder: a preorder where ≤ is total (complete). This is weaker than LinearOrder — it does NOT require antisymmetry or…
  • LinearOrder.toTotalPreorderEvery LinearOrder is a TotalPreorder.

Read back from the Lean

Given a type V with decidable equality and a linear order, a finite set of finite subsets facets : Finset (Finset V), and a field k : Type*, is_cohen_macaulay_two_complex facets k asserts the conjunction of the following five conditions (where verts is the union of all sets in facets, edges is the set of ordered pairs (u, v) ∈ verts × verts with u < v such that {u, v} is a subset of some facet, triangles is the set of triples (u, v, w) ∈ verts × verts × verts with u < v < w such that {u, v, w} ∈ facets, and skel is the simple graph on V where two distinct vertices are adjacent iff they share a facet): 1. facets is non-empty (facets.Nonempty). 2. Every facet has cardinality 3: for every σ ∈ facets, σ.card = 3. 3. The 1-skeleton is connected on the vertex set: for all u ∈ verts and all w ∈ verts, u and w are reachable from each other in skel. 4. The link of every vertex is connected: for every v ∈ verts, and all u, w : V, if {u, v} and {w, v} are both subsets of some facet in facets, and u ≠ v and w ≠ v, then u and w are reachable from each other in the graph on V whose edges are pairs {a, b} with a ≠ b and {a, b, v} ∈ facets. 5. The first homology group over k vanishes ($H_1 = 0$): for every 1-cochain/function c : (V × V) → k, if (a) c e = 0 for all e ∉ edges, and (b) for all x ∈ verts, the divergence $\sum_{(u, v) \in edges} ((\text{if } u = x \text{ then } c(u, v) \text{ else } 0) - (\text{if } v = x \text{ then } c(u, v) \text{ else } 0)) = 0$ (i.e., c is a 1-cycle), then there exists a 2-cochain/function b : (V × (V × V)) → k such that for all e ∈ edges, $c(e) = \sum_{(u, v, w) \in triangles} ((\text{if } e = (v, w) \text{ then } b(u, v, w) \text{ else } 0) - (\text{if } e = (u, w) \text{ then } b(u, v, w) \text{ else } 0) + (\text{if } e = (u, v) \text{ then } b(u, v, w) \text{ else } 0))$ (i.e., c is the boundary $\partial_2 b$).

Written by a model that saw only the Lean, never the English above. If the two disagree, that is worth a challenge.

Lean source view module on GitHub

/-- A finite simplicial complex given by facets is Cohen-Macaulay over a field k if it is pure 2-dimensional, connected, has connected vertex links, and has vanishing first homology. -/
def is_cohen_macaulay_two_complex {V : Type*} [DecidableEq V] [LinearOrder V]
    (facets : Finset (Finset V)) (k : Type*) [Field k] : Prop :=
  let verts := facets.biUnion id
  let edges := (verts ×ˢ verts).filter fun ⟨u, v⟩ => u < v ∧ {u, v} ∈ facets.biUnion Finset.powerset
  let triangles := (verts ×ˢ (verts ×ˢ verts)).filter fun ⟨u, v, w⟩ => u < v ∧ v < w ∧ {u, v, w} ∈ facets
  let skel : SimpleGraph V := SimpleGraph.fromRel (fun u v => ∃ σ ∈ facets, u ∈ σ ∧ v ∈ σ)
  facets.Nonempty ∧
  (∀ σ ∈ facets, σ.card = 3) ∧
  (∀ u ∈ verts, ∀ w ∈ verts, skel.Reachable u w) ∧
  (∀ v ∈ verts, ∀ u w, {u, v} ∈ facets.biUnion Finset.powerset → {w, v} ∈ facets.biUnion Finset.powerset →
    u ≠ v → w ≠ v → (SimpleGraph.fromRel (fun a b => {a, b, v} ∈ facets)).Reachable u w) ∧
  (∀ c : (V × V) → k,
    ((∀ e, e ∉ edges → c e = 0) ∧
     ∀ x ∈ verts, edges.sum (fun ⟨u, v⟩ => (if u = x then c (u, v) else 0) - (if v = x then c (u, v) else 0)) = 0) →
    ∃ b : (V × V × V) → k,
      ∀ e ∈ edges,
        c e = triangles.sum (fun ⟨u, v, w⟩ =>
          (if e = (v, w) then b (u, v, w) else 0) -
          (if e = (u, w) then b (u, v, w) else 0) +
          (if e = (u, v) then b (u, v, w) else 0)))
See it in the graphDiscuss this resultChallenge it