quadratic_neumann_section63_first_index_distinct_centered_case_bound_under_general_sample_bound
Provedcandes-rechtequation-620first-index-distinctlemma-67matrix-completionsection-63source-faithful
This is the centered part of the case in Candes-Recht Section 6.3.
Source location: Candes-Recht 2008, Section 6.3, PDF pp. 31--32. The paper expands
and calls the first resulting sum . It is bounded by Lemma 6.7 with . This node states that bound at the final PDF p. 34 summary scale.
The bound is stated at the Section 6.3 summary scale
The theorem asserts that the centered contribution is at most with probability at least under the full Theorem 1.3 sample lower bound.
Preamble
import Definitions.Def_matrix_completion_neumann open MatrixCompletion
Formal statement
theorem quadratic_neumann_section63_first_index_distinct_centered_case_bound_under_general_sample_bound :
∃ C c : ℝ, 0 < C ∧ 0 < c ∧
∀ C' : ℝ, C ≤ C' →
∀ (β : ℝ), 2 < β →
∀ (n₁ n₂ r m : ℕ) (M : Matrix (Fin n₁) (Fin n₂) ℝ)
(μ₀ μ₁ : ℝ) (S : SVD M r),
0 < n₁ → 0 < n₂ → 0 < r → m ≤ n₁ * n₂ →
1 ≤ μ₀ → 1 ≤ μ₁ →
A0 S μ₀ → A1 S μ₁ →
(m : ℝ) ≥
C' * max (max (μ₁ ^ 2) (Real.sqrt μ₀ * μ₁))
(μ₀ * Real.rpow (↑(max n₁ n₂)) ((1 : ℝ) / 4))
* (↑(max n₁ n₂)) * (r : ℝ) * (β * Real.log (↑(max n₁ n₂))) →
bernoulliEventProb ((m : ℝ) / ((n₁ : ℝ) * (n₂ : ℝ)))
(fun Omega =>
spectralNorm
(quadraticNeumannFirstIndexDistinctCenteredContribution Omega S
((m : ℝ) / ((n₁ : ℝ) * (n₂ : ℝ)))) ≤
(let N : ℝ := ↑(max n₁ n₂)
let R : ℝ := (r : ℝ)
let Mobs : ℝ := (m : ℝ)
let logN : ℝ := Real.log N
C *
((μ₀ ^ 2 * μ₁) *
Real.sqrt ((N * R * (β * logN)) / Mobs) *
((N * R) / Mobs) ^ 2 +
μ₀ ^ 2 * ((N * R) / Mobs) ^ 2 +
Real.sqrt (β * logN) *
Real.rpow ((N * R) / Mobs) ((3 : ℝ) / 2) *
(μ₀ ^ 2 * R) +
Real.rpow
((μ₀ * μ₁ * N * R * (β * logN)) / Mobs)
((3 : ℝ) / 2)))) ≥
1 - c * Real.rpow (↑(max n₁ n₂)) (-β) := by
sorry
Source
Candes, Emmanuel, and Benjamin Recht. "Exact matrix completion via convex optimization." Communications of the ACM 55.6 (2012): 111-119.