Lemma 12.2, second claim — distance is 1-Lipschitz
ProvedModernOnlineLearning.Reductions.lemma_12_2_lipschitzconvex-analysisdistance-functionlipschitzp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-bookp2o-v1
Let be a nonempty closed convex set in a finite-dimensional real normed space. Its distance function satisfies
This controls the dual norm of every subgradient of the distance function and therefore the surrogate gradients.
Preamble
import Mathlib import Definitions.Def_ModernOnlineLearning_Reductions_Setting
Formal statement
namespace ModernOnlineLearning.Reductions
/-- Orabona, Lemma 12.2, second claim, p. 193. -/
theorem lemma_12_2_lipschitz {E : Type*} [NormedAddCommGroup E] [NormedSpace ℝ E]
[FiniteDimensional ℝ E] (V : Set E) (hV : V.Nonempty)
(hclosed : IsClosed V) (hconv : Convex ℝ V) :
LipschitzWith 1 (distance V) := by sorry
end ModernOnlineLearning.Reductions
Source
Orabona, arXiv:1912.13213v10, Lemma 12.2, second claim, p. 193
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.