import OAI.Combinatorics.LowSpaceSubsetSum.Word.WordMain namespace OAI /-! # Worst-case Subset Sum in one-fifth-exponential space The threshold is chosen once from its proved existence theorem. No solution, deficiency statistic, or bit-length exponent is an input. The finite random-word experiment has one-sided correctness, success probability at least two thirds, and bounds on its operation and writable-storage counts on every outcome. -/ noncomputable section namespace LowSpaceSubsetSum def algorithm {n : ℕ} (I : Instance n) (words : WordFinalDecision.Words I) : (Bool × ℕ) × ℕ := WordFinalDecision.run decisionThreshold I words /-- One-sided randomized correctness and all-outcome resource-count bounds. -/ theorem worst_case_subset_sum : (∀ (n : ℕ) (I : Instance n) (words : WordFinalDecision.Words I), (algorithm I words).1.1 = true → Solvable I) ∧ (∀ (n : ℕ) (I : Instance n), Solvable I → (2/3 : ℝ) ≤ (FiniteLaw.uniform : FiniteLaw (WordFinalDecision.Words I)).probability (fun words => (algorithm I words).1.1 = true)) ∧ (∀ c : ℕ, 0 < c → ∀ᶠ n : ℕ in Filter.atTop, ∀ I : Instance n, inputBits I ≤ n^c → ∀ words : WordFinalDecision.Words I, ((algorithm I words).1.2 : ℝ) ≤ (4*(3 : ℝ)^624)*(n : ℝ)^(624*c)*(2 : ℝ)^((1/2 : ℝ)*n) ∧ ((algorithm I words).2 : ℝ) ≤ 2*(inputScale I : ℝ)^15*(2 : ℝ)^((199/1000 : ℝ)*n)) := ⟨fun _ I words => WordFinalDecision.sound decisionThreshold I words, fun _ I => WordFinalDecision.probability decisionThreshold I, WordMain.main_resources decisionThreshold⟩ /-- The writable-storage count has every fixed exponential rate above .199. -/ theorem exponential_space (rho : ℝ) (hrho : (199/1000 : ℝ) < rho) (c : ℕ) (hc : 0 < c) : ∀ᶠ n : ℕ in Filter.atTop, ∀ I : Instance n, inputBits I ≤ n^c → ∀ words : WordFinalDecision.Words I, ((algorithm I words).1.2 : ℝ) ≤ (4*(3 : ℝ)^624)*(n : ℝ)^(624*c)*(2 : ℝ)^((1/2 : ℝ)*n) ∧ ((algorithm I words).2 : ℝ) ≤ (2*(3 : ℝ)^15)*(2 : ℝ)^(rho*n) := WordMain.space_corollary decisionThreshold rho hrho c hc /-- In particular, ordinary one-fifth-exponential writable space. -/ theorem one_fifth_space (c : ℕ) (hc : 0 < c) : ∀ᶠ n : ℕ in Filter.atTop, ∀ I : Instance n, inputBits I ≤ n^c → ∀ words : WordFinalDecision.Words I, ((algorithm I words).1.2 : ℝ) ≤ (4*(3 : ℝ)^624)*(n : ℝ)^(624*c)*(2 : ℝ)^((1/2 : ℝ)*n) ∧ ((algorithm I words).2 : ℝ) ≤ (2*(3 : ℝ)^15)*(2 : ℝ)^((1/5 : ℝ)*n) := exponential_space (1/5) (by norm_num) c hc theorem algorithm_word_width {n : ℕ} (I : Instance n) : wordWidth I = 4*(n+inputBits I+2) := rfl end LowSpaceSubsetSum end end OAI