Documentation

Mathlib.Analysis.Convex.TriangleEquality

Triangle equality for finite sums #

The triangle inequality ‖∑ i ∈ s, v i‖ ≤ ∑ i ∈ s, ‖v i‖ is an equality exactly when the summands pairwise lie on a common closed ray (SameRay). This extends sameRay_iff_norm_add, the two-vector statement in a strictly convex space, to finite families. Equivalently, the nonzero summands all have the same NormedSpace.normalize, i.e. every summand is a nonnegative real multiple of a single vector.

Main statements #

Tags #

triangle inequality, triangle equality, same ray, strictly convex space

theorem norm_sum_eq_of_pairwise_sameRay {ι : Type u_1} {s : Finset ι} {E : Type u_2} [SeminormedAddCommGroup E] [NormedSpace ℝ E] {v : ι → E} (hp : (↑s).Pairwise (Function.onFun (SameRay ℝ) v)) :
‖∑ i ∈ s, v i‖ = ∑ i ∈ s, ‖v i‖

If the summands pairwise lie on a common closed ray, the norm of their sum is the sum of their norms.

theorem sum_ne_zero_of_pairwise_sameRay {ι : Type u_1} {s : Finset ι} {i : ι} {E : Type u_2} [NormedAddCommGroup E] [NormedSpace ℝ E] {v : ι → E} (hp : (↑s).Pairwise (Function.onFun (SameRay ℝ) v)) (hi : i ∈ s) (hvi : v i ≠ 0) :
∑ j ∈ s, v j ≠ 0
theorem norm_sum_eq_iff_pairwise_sameRay {ι : Type u_1} {s : Finset ι} {E : Type u_2} [NormedAddCommGroup E] [NormedSpace ℝ E] [StrictConvexSpace ℝ E] {v : ι → E} :
‖∑ i ∈ s, v i‖ = ∑ i ∈ s, ‖v i‖ ↔ (↑s).Pairwise (Function.onFun (SameRay ℝ) v)

Triangle equality for a finite sum: the norm of the sum equals the sum of the norms exactly when the summands pairwise lie on a common closed ray.

theorem normalize_eq_of_pairwise_sameRay {ι : Type u_1} {s : Finset ι} {i : ι} {E : Type u_2} [NormedAddCommGroup E] [NormedSpace ℝ E] {v : ι → E} (hp : (↑s).Pairwise (Function.onFun (SameRay ℝ) v)) (hi : i ∈ s) (hvi : v i ≠ 0) :

If the summands pairwise lie on a common closed ray and one of them is nonzero, then it has the same normalization as the sum.

theorem norm_sum_eq_iff_exists_smul {ι : Type u_1} {s : Finset ι} {E : Type u_2} [NormedAddCommGroup E] [NormedSpace ℝ E] [StrictConvexSpace ℝ E] {v : ι → E} :
‖∑ i ∈ s, v i‖ = ∑ i ∈ s, ‖v i‖ ↔ ∃ (c : E), ∀ i ∈ s, v i = ‖v i‖ • c

Triangle equality: the norm of a finite sum equals the sum of the norms exactly when every summand is a nonnegative real multiple of a single vector.