Documentation

FirstOrderMethodsOptimization_Beck_2017.Chap02.Example_2_9

theorem support_function_unit_ball_eq_dualNorm {E : Type u} [NormedAddCommGroup E] [NormedSpace ℝ E] [FiniteDimensional ℝ E] (y : Module.Dual ℝ E) :
(σ_ (Metric.closedBall 0 1)) y = ↑(dualNorm y)

Example 2.9 (1): the support function of the closed unit ball of a normed space is the dual norm.

theorem support_function_lp_unit_ball_eq_conjugate_lp_norm {n : ℕ} {p q : ENNReal} (hpq : p.HolderConjugate q) (y : Fin n → ℝ) :
(σ_ {x : Fin n → ℝ | ‖WithLp.toLp p x‖ ≤ 1}) ((dotProductBilin ℝ ℝ) y) = ↑‖WithLp.toLp q y‖

Example 2.9 (2): in ℝ^n with the l_p norm, the support function of the closed unit ball is the conjugate l_q norm.

theorem support_function_posDef_unit_ball_eq_sqrt_dotProduct_inv_mulVec {n : ℕ} (Q : Matrix (Fin n) (Fin n) ℝ) (hQ : Q.PosDef) (y : Fin n → ℝ) :
(σ_ {x : Fin n → ℝ | Q.qNorm hQ x ≤ 1}) ((dotProductBilin ℝ ℝ) y) = ↑√(y ⬝ᵥ Q⁻¹.mulVec y)

Example 2.9 (3): for the norm induced by a positive definite matrix Q, the support function of the closed unit ball is the Q⁻¹-norm, written here as √(yᵀ Q⁻¹ y).