Proof of Theorem 2.1, p. 247 — every x_ε = (1 − ε)x* + εx ∈ X_ε satisfies f(x_ε) ≤ f(x*) + 2εB
ProvedConvexOptAlg.CenterGravity.thm_2_1_value_scaled_copycenter-of-gravityconvex-optimizationp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-bookp2o-v1
Let be a convex body, continuous and convex, a minimizer of on , and . For every the point of satisfies
Points of the shrunk copy are -optimal; this is the last inequality of the proof of Theorem 2.1.
Formalization Note is required on only; values of outside play no role. The standing assumptions of Chapter 2 are hypotheses.
Preamble
import Mathlib import Definitions.Def_ConvexOptAlg_CenterGravity_Defs open MeasureTheory open scoped InnerProductSpace
Formal statement
namespace ConvexOptAlg.CenterGravity
/-- Bubeck, arXiv:1405.4980v2, proof of Theorem 2.1, p. 247: for `ε ∈ [0, 1]` and every point
`x_ε = (1 - ε)x* + εx` of `X_ε` (`x ∈ X`), convexity of `f` gives `f(x_ε) ≤ f(x*) + 2εB`. Standing
assumptions of Ch. 2: `X` a convex body, `f : X → [-B, B]` continuous and convex, `x*` a minimizer. -/
theorem thm_2_1_value_scaled_copy {n : ℕ} {X : Set (EuclideanSpace ℝ (Fin n))} (hX : IsConvexBody X)
{f : EuclideanSpace ℝ (Fin n) → ℝ} {B : ℝ} (hfB : ∀ x ∈ X, |f x| ≤ B)
(hfc : ContinuousOn f X) (hfconv : ConvexOn ℝ X f)
{xstar : EuclideanSpace ℝ (Fin n)} (hxstar : xstar ∈ X) (hmin : ∀ y ∈ X, f xstar ≤ f y)
(ε : ℝ) (hε0 : 0 ≤ ε) (hε1 : ε ≤ 1) (x : EuclideanSpace ℝ (Fin n)) (hx : x ∈ X) :
f ((1 - ε) • xstar + ε • x) ≤ f xstar + 2 * ε * B := by sorry
end ConvexOptAlg.CenterGravity
Source
Bubeck, arXiv:1405.4980v2, proof of Theorem 2.1, p. 247
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.