linear_flattening_testable_succ_dim
theoremverified
Testability is monotone in the tensor dimension at fixed order: if a linear flattening on (ℂⁿ)⊗³ has potency greater than r, then so does one on (ℂ^{n+1})⊗³, obtained by restricting a tensor to its first n coordinates in each mode.
Statement
theorem linear_flattening_testable_succ_dim {n r : ℕ}
(h : linear_flattening_testable n r) : linear_flattening_testable (n + 1) r- Source
- step towards linear_flattening_testable_even_step
- Verified
- 23 Oct 2026
- Axioms
Classical.choiceQuot.soundpropext- Built on 3
- linear_flattening_testableThe secant variety σ_r((ℂⁿ)⊗³) is testable by the linear flattening method if there exists a linear flattening L : (ℂⁿ)⊗³ → Mat_{u×v}(ℂ)…
- linear_flattening_potency_gtA linear flattening L : ℂⁿ ⊗ ℂⁿ ⊗ ℂⁿ → Mat_{u×v}(ℂ) has potency greater than r: some tensor T has rank L(T) > r · ρ(L), where ρ(L) =…
- tensor3_pureThe pure tensor x ⊗ y ⊗ z in ℂⁿ ⊗ ℂⁿ ⊗ ℂⁿ, with entries x_i y_j z_k.
- Used by 1
- linear_flattening_testable_mono_dimTestability is monotone in the tensor dimension: if σ_r((ℂⁿ)⊗³) is testable by linear flattenings and n ≤ m, then so is σ_r((ℂᵐ)⊗³).
The proof, explained
Summary of the proof.
Linearity of restriction
Linearity holds by pointwise definition.
map_add' A B := by ext i j k; rfl map_smul' c A := by ext i j k; rfl
Proof of theorem
By padding and restricting.
rcases h with ⟨u, v, L, T, hT⟩ refine ⟨u, v, L.comp linear_flattening_testable_succ_dim_restrict, linear_flattening_testable_succ_dim_pad T, ?_⟩ intro x y z have hq_pure : linear_flattening_testable_succ_dim_restrict (tensor3_pure x y z) = tensor3_pure (fun i => x i.castSucc) (fun i => y i.castSucc) (fun i => z i.castSucc) := by ext i j k simp [linear_flattening_testable_succ_dim_restrict, tensor3_pure] have hq_pad : linear_flattening_testable_succ_dim_restrict (linear_flattening_testable_succ_dim_pad T) = T := by ext i j k simp [linear_flattening_testable_succ_dim_restrict, linear_flattening_testable_succ_dim_pad, Fin.lastCases_castSucc] change r * (L (linear_flattening_testable_succ_dim_restrict (tensor3_pure x y z))).rank < (L (linear_flattening_testable_succ_dim_restrict (linear_flattening_testable_succ_dim_pad T))).rank rw [hq_pure, hq_pad] exact hT _ _ _
Written by a model (gemini-3.8-flash) from the Lean proof and the goal states Lean reported at each step. Lean verified the proof; this explanation is a reading aid and is not itself machine-checked.
Lean source view module on GitHub
Choose a tactic — a marked line — to see the goals Lean had before and after it, as in Lean’s infoview. Step with ← → or j k.
def linear_flattening_testable_succ_dim_restrict {n : ℕ} : (Fin (n + 1) → Fin (n + 1) → Fin (n + 1) → ℂ) →ₗ[ℂ] (Fin n → Fin n → Fin n → ℂ) where toFun A i j k := A i.castSucc j.castSucc k.castSucc map_add' A B := by ext i j k; rfl map_smul' c A := by ext i j k; rfldef linear_flattening_testable_succ_dim_pad {n : ℕ} (T : Fin n → Fin n → Fin n → ℂ) : Fin (n + 1) → Fin (n + 1) → Fin (n + 1) → ℂ := fun i j k => Fin.lastCases (0 : ℂ) (fun a => Fin.lastCases (0 : ℂ) (fun b => Fin.lastCases (0 : ℂ) (fun c => T a b c) k) j) i/-- Padding a flattening by a coordinate preserves its potency at fixed order; note it never raises the order, since every pure-tensor image of the small flattening is already a pure-tensor image of the padded one. -/theorem linear_flattening_testable_succ_dim {n r : ℕ} (h : linear_flattening_testable n r) : linear_flattening_testable (n + 1) r := by rcases h with ⟨u, v, L, T, hT⟩ refine ⟨u, v, L.comp linear_flattening_testable_succ_dim_restrict, linear_flattening_testable_succ_dim_pad T, ?_⟩ intro x y z have hq_pure : linear_flattening_testable_succ_dim_restrict (tensor3_pure x y z) = tensor3_pure (fun i => x i.castSucc) (fun i => y i.castSucc) (fun i => z i.castSucc) := by ext i j k simp [linear_flattening_testable_succ_dim_restrict, tensor3_pure] have hq_pad : linear_flattening_testable_succ_dim_restrict (linear_flattening_testable_succ_dim_pad T) = T := by ext i j k simp [linear_flattening_testable_succ_dim_restrict, linear_flattening_testable_succ_dim_pad, Fin.lastCases_castSucc] change r * (L (linear_flattening_testable_succ_dim_restrict (tensor3_pure x y z))).rank < (L (linear_flattening_testable_succ_dim_restrict (linear_flattening_testable_succ_dim_pad T))).rank rw [hq_pure, hq_pad] exact hT _ _ _