L, R and M = L⁻¹R for (3.10), p. 12 — the 5 × 5 matrices of the example, entry by entry
ProvedExtADMM.Diverge.matrices_3_10admmmatrix-computationp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1
For the data
the matrices of (3.6) and of (3.7) are
, and
This pins every printed entry of the example's iteration matrix, which is the object of the spectral and divergence statements that follow.
Formalization Note. are the columns of (3.10): , , . The determinant is stated so that is the genuine inverse rather than Mathlib's junk value for singular matrices.
Preamble
import Mathlib import Definitions.Def_ExtADMM_Diverge_Setting open Matrix Filter Topology
Formal statement
namespace ExtADMM.Diverge
/-- p. 12. At the data (3.10), the matrices `L` of (3.6) and `R` of (3.7) are the printed
`L310` and `R310`; `L310` is nonsingular (`det = 54`) and `M = L⁻¹R` is the printed `M310`. -/
theorem matrices_3_10 :
Lmat a310_2 a310_3 = L310 ∧ Rmat a310_1 a310_2 a310_3 = R310 ∧
L310.det = 54 ∧ L310⁻¹ * R310 = M310 := by sorry
end ExtADMM.Diverge
Source
Chen, He, Ye & Yuan, The direct extension of ADMM for multi-block convex minimization problems is not necessarily convergent, Math. Program., DOI 10.1007/s10107-014-0826-5 (authors' version of January 22, 2014), p. 12, (3.10) and the matrices L, R, M = L⁻¹R
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.