Lemma 8.7.3 — boundary characterization of the unique smallest enclosing ball
ProvedMatousekLP.SmallestBall.unique_smallest_ball_iffdiscrete-geometryp2o-batch-b23bp2o-gran-per-chapterp2o-plan-bookp2o-v1smallest-enclosing-ball
Let be the closed ball in with center and radius , and let be points on the boundary of , i.e. for all . Then the following two statements are equivalent.
- is the unique smallest enclosing ball of : it contains , every ball containing has radius at least , and every ball containing with radius at most has center .
- For every there is an index with
Condition 2 says that no hyperplane strictly separates from . The lemma characterizes smallest enclosing balls by their boundary points and is the geometric half of the proof of Theorem 8.7.4.
Formalization Note The points are indexed by Fin k (0-based) and is their range; is the Euclidean inner product. For both statements are false (a ball of negative radius, which is empty, encloses ; and no index exists), so no hypothesis is needed.
Preamble
import Mathlib import Definitions.Def_MatousekLP_SmallestBall_Basic open scoped RealInnerProductSpace
Formal statement
namespace MatousekLP.SmallestBall
/-- Lemma 8.7.3 (Matoušek & Gärtner, p. 188). Let `S = {s₁, …, s_k} ⊆ ℝ^d` lie on the boundary of
the ball `B` with center `s*` and radius `r` (so `‖sⱼ − s*‖ = r` for all `j`). Then `B` is the
unique smallest enclosing ball of `S` iff for every `u ∈ ℝ^d` there is an index `j` with
`uᵀ(sⱼ − s*) ≤ 0`. -/
theorem unique_smallest_ball_iff {d k : ℕ} (s : Fin k → EuclideanSpace ℝ (Fin d))
(sstar : EuclideanSpace ℝ (Fin d)) (r : ℝ) (hr : 0 ≤ r)
(hbd : ∀ j, dist (s j) sstar = r) :
IsUniqueSmallestEnclosingBall (Set.range s) sstar r ↔
∀ u : EuclideanSpace ℝ (Fin d), ∃ j : Fin k, ⟪u, s j - sstar⟫ ≤ 0 := by sorry
end MatousekLP.SmallestBall
Source
Matoušek & Gärtner, Understanding and Using Linear Programming, Springer 2007, p. 188, Lemma 8.7.3
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.