Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 4, p. 9 — BLP+Affine solves PCSP(A, B) iff Pol(A, B) has block-symmetric polymorphisms of arbitrarily high width iff (2L+1)-ary ones with blocks L, L+1

Proved
PCSPBLPAff.Characterization.theorem_4

by mikedeng1 · Oct 7, 2026 · Mathlib 0df444a (Lean v4.33.1)

affine-relaxationbasic-lpblock-symmetricp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1polymorphismpromise-csp

Let (A,B)(\mathbf A,\mathbf B)(A,B) be a promise template on finite domains AAA, BBB. The following are equivalent:

  1. the BLP+Affine algorithm correctly solves PCSP-Decision(A,B)\mathrm{PCSP\text{-}Decision}(\mathbf A,\mathbf B)PCSP-Decision(A,B), i.e. it accepts every instance satisfiable in A\mathbf AA and rejects every instance unsatisfiable in B\mathbf BB;
  2. Pol(A,B)\mathrm{Pol}(\mathbf A,\mathbf B)Pol(A,B) has block-symmetric polymorphisms of arbitrarily high width: for every N∈NN\in\mathbb NN∈N some polymorphism fff is invariant under permutations within each block of a partition of its coordinates into at least one block, every block having at least NNN elements;
  3. for every L∈NL\in\mathbb NL∈N, Pol(A,B)\mathrm{Pol}(\mathbf A,\mathbf B)Pol(A,B) has a block-symmetric polymorphism of arity 2L+12L+12L+1 with two symmetric blocks of variables of sizes LLL and L+1L+1L+1.
BLP+Affine solves PCSP(A,B)  ⟺  sup⁡f∈Pol(A,B)width(f)=∞  ⟺  ∀L ∃f∈Pol(2L+1) symmetric on blocks of sizes L, L+1.\text{BLP+Affine solves }\mathrm{PCSP}(\mathbf A,\mathbf B)\iff \sup_{f\in\mathrm{Pol}(\mathbf A,\mathbf B)}\mathrm{width}(f)=\infty\iff \forall L\ \exists f\in\mathrm{Pol}^{(2L+1)}\ \text{symmetric on blocks of sizes }L,\,L+1.BLP+Affine solves PCSP(A,B)⟺f∈Pol(A,B)sup​width(f)=∞⟺∀L ∃f∈Pol(2L+1) symmetric on blocks of sizes L,L+1.

This is the paper's exact characterization of the power of the combined Basic LP and affine relaxation for promise CSPs.

Formalization Note The algorithm is formalized through its acceptance condition (see the definitions file); the polynomial-time claims are not formalized. In item 3 the two blocks are given by a labelling β:[2L+1]→{0,1}\beta:[2L+1]\to\{0,1\}β:[2L+1]→{0,1} with ∣β−1(0)∣=L|\beta^{-1}(0)|=L∣β−1(0)∣=L and ∣β−1(1)∣=L+1|\beta^{-1}(1)|=L+1∣β−1(1)∣=L+1.

Preamble
import Mathlib
import Definitions.Def_PCSPBLPAff_Characterization_Setting
Formal statement
namespace PCSPBLPAff.Characterization

/-- Theorem 4, p. 9: for a promise template `(𝔸, 𝔹)` the following are equivalent:
(1) the BLP+Affine algorithm correctly solves `PCSP-Decision(𝔸, 𝔹)`;
(2) `Pol(𝔸, 𝔹)` has block-symmetric polymorphisms of arbitrarily high width;
(3) for every `L ∈ ℕ`, `Pol(𝔸, 𝔹)` has a block-symmetric polymorphism of arity `2L + 1` with
two symmetric blocks of variables of size `L` and `L + 1`. -/
theorem theorem_4 {τ : Type} {ar : τ → ℕ} {A B : Type} [Fintype A] [DecidableEq A] [Fintype B]
    (𝔸 : PCSPBLPAff.Symmetric.RelStruct τ ar A) (𝔹 : PCSPBLPAff.Symmetric.RelStruct τ ar B) (hT : PCSPBLPAff.Symmetric.IsPromiseTemplate 𝔸 𝔹) :
    List.TFAE
      [ PCSPBLPAff.Symmetric.CorrectlySolves 𝔸 𝔹,
        ∀ N : ℕ, ∃ (L : ℕ) (f : (Fin L → A) → B), PCSPBLPAff.Symmetric.IsPolymorphism 𝔸 𝔹 f ∧ PCSPBLPAff.Symmetric.HasWidthAtLeast f N,
        ∀ L : ℕ, ∃ f : (Fin (2 * L + 1) → A) → B, PCSPBLPAff.Symmetric.IsPolymorphism 𝔸 𝔹 f ∧
          ∃ β : Fin (2 * L + 1) → Fin 2,
            (Finset.univ.filter (fun l => β l = 0)).card = L ∧
            (Finset.univ.filter (fun l => β l = 1)).card = L + 1 ∧
            PCSPBLPAff.Symmetric.IsBlockSymmetricFor β f ] := by sorry

end PCSPBLPAff.Characterization
Source
arXiv:1907.04383v3, Theorem 4, p. 9
Human review
  • Endorsed by Shuze Chen · Oct 8, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 8, 2026

    Confirmed by the mission captain (proposal self-audit).

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me