Documentation

LeanPool.SNumbers.BasicResults.KadetsSnobar

The Kadets–Snobar projection theorem #

Every finite-dimensional subspace of a Banach space is the range of a bounded projection whose norm is at most the square root of its dimension. This is a classical fact of Banach-space geometry (Kadets–Snobar 1971; Pietsch, Eigenvalues and s-numbers 1.5.5).

The theorem is reduced to the John's-ellipsoid development in BasicResults.John: John.exists_projection supplies the projection with ‖P‖ ≤ √(dim V), and we weaken dim V to n; the underlying input is John.john_decomposition, the John decomposition of identity.

theorem SNumbers.exists_projection_range_eq_of_rank_le {𝕜 : Type u} [RCLike 𝕜] {Y : Type u} [NormedAddCommGroup Y] [NormedSpace 𝕜 Y] {V : Submodule 𝕜 Y} {n : ℕ} (hV_rank : Module.rank 𝕜 ↥V ≤ ↑n) :
∃ (P : Y →L[𝕜] Y), P ∘SL P = P ∧ (↑P).range = V ∧ ‖P‖ ≤ √↑n

Kadets–Snobar theorem. Every finite-dimensional subspace V of a Banach space Y of dimension at most n is the range of a bounded projection P : Y →L[𝕜] Y (P ∘ P = P) with operator norm ‖P‖ ≤ √n.