AFTD Auto-Formalizing Theoretical Domains

tsp_short

def

For a vertex v in a finite set V with decidable equality and edge weights ℓ : V → V → ℝ, tsp_short ℓ v is the minimum of ℓ(v, u) over all u in Finset.univ.erase v if that set is nonempty, and 0 otherwise.

Statement

noncomputable def tsp_short {V : Type*} [Fintype V] [DecidableEq V] (ℓ : V → V → ℝ) (v : V) : ℝ
Defined in
arXiv:2611.09421 Section 2.1
Verified
10 Nov 2026
Axioms
none listed

Read back from the Lean

For a finite type V with decidable equality, a cost/distance function ℓ : V → V → ℝ, and an element v : V: tsp_short ℓ v is the minimum value of ℓ v w over all w ∈ V with w ≠ v, provided at least one such w exists; if no such w exists (i.e., V contains only v), it evaluates to 0.

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

/-- For a vertex v in a complete edge-weighted graph, short(v) is the minimum length of an edge incident to v. -/
noncomputable def tsp_short {V : Type*} [Fintype V] [DecidableEq V] (ℓ : V → V → ℝ) (v : V) : ℝ := if h : (Finset.univ.erase v).Nonempty then ((Finset.univ.erase v).image (ℓ v)).min' (Finset.Nonempty.image h _) else 0
See it in the graphDiscuss this resultChallenge it