A strongly convex gradient step contracts distances
AutoSamplingTheory.TechnicalLemmas.Analysis.GradientDescentContraction.gradient_step_contraction · theorem · Teaching coverage
Statement
For every pair x,y, the actual step T satisfies the displayed contraction estimate under the stated global curvature and step conditions.
All objects and hypotheses
- E is a complete real inner-product space, f:E→ℝ is C¹, and all point quantifiers range over E. The gradient is the genuine Riesz representative of its Fréchet derivative.
- The real parameters α, β and h are nonnegative and βh≤1.
- f is α-strongly convex on all E: the convex chord inequality has the subtraction (α/2)t(1−t)‖y−x‖² for t∈[0,1]. The displayed β quadratic upper model holds for all x,y.
- Write T(x)=x−h∇f(x) and r=√(1−αh), with the nonnegative real square root (zero on negative arguments).
Mathematical proof
1. Use the two compiled curvature inequalities
For arbitrary x,y define v=y−x, d=∇f(y)−∇f(x) and p=⟨d,v⟩. Strong convexity with α≥0 supplies convexity and the lower pairing bound. Cocoercivity supplies the upper gradient-square bound.
Corresponding Lean step
StrongConvexFirstOrder.gradient_inner_lower_bound_of_strongConvexOn; UniformConvexOn.convexOn; ConvexSmoothGradient.gradient_cocoercive.
2. Expand one actual step
Since h≥0 and βh≤1, multiplying the cocoercivity bound by h² gives h²‖d‖²≤h²βp≤hp. Insert this into the exact squared-norm expansion, then use hp≥αh‖v‖².
Corresponding Lean step
norm_sub_sq_real, inner_smul_right, norm_smul; three mul_le_mul_of_nonneg_left applications and nlinarith.
3. Take a square root without adding compatibility
The squared norm is nonnegative. Taking square roots and factoring the nonnegative factor ‖v‖² yields the asserted distance bound. If 1−αh is negative and v≠0, the squared estimate is impossible; hence no nontrivial negative-coefficient case is hidden.
Corresponding Lean step
Real.le_sqrt_of_sq_le; Real.sqrt_mul' with sq_nonneg; Real.sqrt_sq with norm_nonneg.
Lean statement · gradient_step_contraction
For every pair x,y, the actual step T satisfies the displayed contraction estimate under the stated global curvature and step conditions.
Braces mark parameters Lean can infer; square brackets request structures such as a measurable space or probability measure. Named hypotheses are mathematical premises, not facts established by this declaration. Section parameters are described in the mathematical hypotheses above; the module link retains their exact source context.
theorem gradient_step_contraction {f : E → ℝ} {α β h : ℝ}
(hf : ContDiff ℝ 1 f) (hsc : StrongConvexOn univ α f)
(hα : 0 ≤ α) (hβ : 0 ≤ β) (hh : 0 ≤ h) (hstep : β * h ≤ 1)
(hu : ∀ x y, f y ≤ f x + inner ℝ (gradient f x) (y - x) + β / 2 * ‖y - x‖ ^ 2)
(x y : E) :
‖(y - h • gradient f y) - (x - h • gradient f x)‖ ≤
Real.sqrt (1 - α * h) * ‖y - x‖Lean proof · gradient_step_contraction
For arbitrary x,y define v=y−x, d=∇f(y)−∇f(x) and p=⟨d,v⟩. Strong convexity with α≥0 supplies convexity and the lower pairing bound. Cocoercivity supplies the upper gradient-square bound. Since h≥0 and βh≤1, multiplying the cocoercivity bound by h² gives h²‖d‖²≤h²βp≤hp. Insert this into the exact squared-norm expansion, then use hp≥αh‖v‖². The squared norm is nonnegative. Taking square roots and factoring the nonnegative factor ‖v‖² yields the asserted distance bound. If 1−αh is negative and v≠0, the squared estimate is impossible; hence no nontrivial negative-coefficient case is hidden.
Braces mark parameters Lean can infer; square brackets request structures such as a measurable space or probability measure. Named hypotheses are mathematical premises, not facts established by this declaration. Section parameters are described in the mathematical hypotheses above; the module link retains their exact source context.
theorem gradient_step_contraction {f : E → ℝ} {α β h : ℝ}
(hf : ContDiff ℝ 1 f) (hsc : StrongConvexOn univ α f)
(hα : 0 ≤ α) (hβ : 0 ≤ β) (hh : 0 ≤ h) (hstep : β * h ≤ 1)
(hu : ∀ x y, f y ≤ f x + inner ℝ (gradient f x) (y - x) + β / 2 * ‖y - x‖ ^ 2)
(x y : E) :
‖(y - h • gradient f y) - (x - h • gradient f x)‖ ≤
Real.sqrt (1 - α * h) * ‖y - x‖ := by
have hc : ConvexOn ℝ univ f := hsc.convexOn (by intro r; positivity)
have hm := StrongConvexFirstOrder.gradient_inner_lower_bound_of_strongConvexOn
hsc (fun z _ => (hf.differentiable_one z).hasGradientAt) (mem_univ x) (mem_univ y)
have hg := ConvexSmoothGradient.gradient_cocoercive hf hc hβ hu x y
have hp : 0 ≤ inner ℝ (gradient f y - gradient f x) (y - x) :=
le_trans (mul_nonneg hα (sq_nonneg _)) hm
have hg' := mul_le_mul_of_nonneg_left hg (sq_nonneg h)
have hs := mul_le_mul_of_nonneg_left hstep (mul_nonneg hh hp)
have hm' := mul_le_mul_of_nonneg_left hm hh
have hv : (y - h • gradient f y) - (x - h • gradient f x) =
(y - x) - h • (gradient f y - gradient f x) := by rw [smul_sub]; abel
have hb : ‖(y - h • gradient f y) - (x - h • gradient f x)‖ ^ 2 ≤
(1 - α * h) * ‖y - x‖ ^ 2 := by
rw [hv, norm_sub_sq_real, inner_smul_right,
real_inner_comm (gradient f y - gradient f x), norm_smul,
Real.norm_eq_abs, mul_pow, sq_abs]
nlinarith
simpa only [Real.sqrt_mul' _ (sq_nonneg _), Real.sqrt_sq (norm_nonneg _)] using
Real.le_sqrt_of_sq_le hb
/-- The actual Nth gradient-descent iterate has geometric, then exponential,
distance control about a supplied global minimizer. -/Scope and omitted-condition boundaries
- C¹ on a complete real inner-product space is an explicit extension of the section’s C² Euclidean setting. The source reciprocal restriction is recovered for β>0; the division-free statement also includes β=0 and h=0. If 1−αh<0, the squared bound forces every pairwise distance to vanish, so the norm result remains valid on a singleton space without an unjustified α≤β assumption. No constrained-domain, Riemannian or stochastic method is covered; this is shared optimization background, not companion-paper progress.
- The source prints h≤1/β without stating h≥0. Nonnegative step is essential: f(t)=t²/2, α=β=1 and h=−1 gives T(t)=2t, contradicting the claimed factor √2 for distinct points. The reciprocal formula additionally needs β>0. The stray gradient before y−x in the middle proof display is a transcription typo, resolved by the preceding expansion and Proposition1.6.
- The source specializes to h=1/β and κ=β/α. For 0<α≤β, r^N=(1−1/κ)^(N/2) and the exponential factor is exp(−N/(2κ)). This unit does not formalize the subsequent logarithmic iteration-complexity statement or the improved step of Exercise3.2.
Source and reuse
ASTIS parents called
AutoSamplingTheory.TechnicalLemmas.Analysis.StrongConvexFirstOrder.gradient_inner_lower_bound_of_strongConvexOnAutoSamplingTheory.TechnicalLemmas.Analysis.ConvexSmoothGradient.gradient_cocoercive
Mathlib API called (external library)
- UniformConvexOn.convexOn; ContDiff.differentiable_one; DifferentiableAt.hasGradientAt
- norm_sub_sq_real; inner_smul_right; norm_smul; mul_le_mul_of_nonneg_left; Real.le_sqrt_of_sq_le; Real.sqrt_mul'; Real.sqrt_sq
Mathematical sources
- Sinho Chewi, Theorem3.3 and its proof — ASTIS-authored restatement and proof explanation; necessary step condition exposed.
- Pinned Optlib gradient_method_strong_convex — Related stronger interpolation route, Apache-2.0; external reference only, not an imported dependency.
- Reciprocal step, one-step termination and zero-modulus consumers — Compiled calls of the actual new interfaces.
ASTIS prose is not a quotation or a source-equivalence certificate. Definitions and aliases are explained as constructions, not counted as new mathematical proofs.