(2.20) — Ψ(x^ag_k) ≤ Ψ(x^md_k) − β_k‖∇Ψ(x^md_k)‖² + (L_Ψβ_k²/2)‖∇Ψ(x^md_k)‖²
ProvedNonconvexAG.Smooth.eq_2_20accelerated-gradientnonconvex-optimizationp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1
Let be differentiable with -Lipschitz gradient, , and let be generated by Algorithm 1 from with step sizes , (), , . Then for every ,
It is the descent of the gradient step (2.4) and opens the proof of Theorem 1 b).
Formalization Note The statement is the outer ends of the chain (2.20).
Preamble
import Mathlib import Definitions.Def_ConvexOptAlg_SmoothGD_Defs import Definitions.Def_NonconvexAG_Smooth_AGRun
Formal statement
namespace NonconvexAG.Smooth
/-- (2.20), p. 6 (outer ends of the chain). -/
theorem eq_2_20 {n : ℕ} (Ψ : E n → ℝ) (g : E n → E n) (LΨ : ℝ)
(hΨ : ConvexOptAlg.SmoothGD.IsBetaSmooth Ψ g LΨ) (hL : 0 < LΨ)
(α β lam : ℕ → ℝ) (hstep : AGStepsizes α β lam) (x0 : E n) :
∀ k, 1 ≤ k →
Ψ (xagSeq g α β lam x0 k) ≤ Ψ (xmdSeq g α β lam x0 k) - β k * ‖g (xmdSeq g α β lam x0 k)‖ ^ 2
+ LΨ * β k ^ 2 / 2 * ‖g (xmdSeq g α β lam x0 k)‖ ^ 2 := by sorry
end NonconvexAG.Smooth
Source
Ghadimi & Lan, Accelerated Gradient Methods for Nonconvex Nonlinear and Stochastic Programming, arXiv:1310.3787v1, p. 6, (2.20)
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.