adjacency_lambda2
For any finite vertex type V and any simple graph G on V, the eigenvalue at index 1 of the real adjacency matrix (G.isHermitian_adjMatrix ℝ).eigenvalues₀ when 2 ≤ Fintype.card V (which is the second largest eigenvalue λ₂ of the adjacency matrix, as Mathlib's Matrix.IsHermitian.eigenvalues₀ is antitone: index 0 is the largest and index 1 is the second largest), defaulting to 0 if Fintype.card V < 2.
Statement
noncomputable def adjacency_lambda2 {V : Type*} [Fintype V] (G : SimpleGraph V) : ℝ
- Defined in
- arXiv:2611.03879, Section 1
- Verified
- 05 Nov 2026
- Axioms
- none listed
- Used by 2
- bipartite_complement_adj_spectrum_and_lambda2Let H be an s-regular bipartite simple graph whose two parts V and W each have size m ≥ 2, and let G = K_{V,W} \ E(H) be its bipartite…
- adjacency_lambda2_eq_of_charpoly_rootsFor a simple graph G on a finite vertex set, if the roots of the characteristic polynomial of its real adjacency matrix are a, x and a…
Read back from the Lean
For an arbitrary type V with a Fintype V instance (i.e., V is finite) and a simple graph G : SimpleGraph V, adjacency_lambda2 G is defined as a real number (ℝ):
- If the number of vertices in V is at least 2 (h : 2 ≤ Fintype.card V), it evaluates to (G.isHermitian_adjMatrix ℝ).eigenvalues₀ ⟨1, h⟩, which is the eigenvalue at index 1 (the second element under 0-based indexing of Fin (Fintype.card V)) of the Hermitian real adjacency matrix of G.
- Otherwise (if Fintype.card V < 2, i.e., V has 0 or 1 vertices), 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
/-- The second largest eigenvalue of the real adjacency matrix of a finite simple graph. -/ noncomputable def adjacency_lambda2 {V : Type*} [Fintype V] (G : SimpleGraph V) : ℝ := open scoped Classical in if h : 2 ≤ Fintype.card V then (G.isHermitian_adjMatrix ℝ).eigenvalues₀ ⟨1, h⟩ else 0