Appendix B, p. 40 — ‖P⁸‖² < 1/2 in the induced 2-norm
ProvedIQCAlg.HeavyBall.P8_normheavy-ballmatrixoperator-normp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1
Let and let be the operator norm induced by the Euclidean norm on . Then
(Numerically .)
Consequently is a contraction, which turns the linear perturbation recursion into a monotone decrease of the error every eight steps.
Formalization Note The norm is Mathlib's operator norm on matrices, activated by open scoped Matrix.Norms.L2Operator; without it, Matrix carries no norm (and the entrywise sup norm, which is not the page's norm, is the other common choice).
Preamble
import Mathlib import Definitions.Def_IQCAlg_HeavyBall_Setting open scoped Matrix.Norms.L2Operator
Formal statement
namespace IQCAlg.HeavyBall /-- p. 40: `‖P⁸‖² < 1/2` for the induced 2-norm (Mathlib's `ℓ²` operator norm on matrices). -/ theorem P8_norm : ‖Pm ^ 8‖ ^ 2 < 1 / 2 := by sorry end IQCAlg.HeavyBall
Source
Lessard, Recht & Packard, Analysis and Design of Optimization Algorithms via Integral Quadratic Constraints, arXiv:1408.3595v7, p. 40, ‖P⁸‖² ≈ 0.46044 < 1/2
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.