theorem
corollary_1_7_upper_tail
{Ω : Type u_1}
[MeasurableSpace Ω]
{μ : MeasureTheory.Measure Ω}
[MeasureTheory.IsProbabilityMeasure μ]
{n : ℕ}
{X : Fin n → Ω → ℝ}
{σsq : ℝ}
(hX_sg : ∀ (i : Fin n), IsSubGaussian (X i) σsq μ)
(hX_indep : ProbabilityTheory.iIndepFun X μ)
(hX_meas : ∀ (i : Fin n), Measurable (X i))
(a : Fin n → ℝ)
(t : ℝ)
(ht : 0 < t)
:
Corollary 1.7 — Upper tail bound for linear combinations.
For independent sub-Gaussian random variables Xᵢ ~ subG(σ²) and any a : Fin n → ℝ,
P[∑ᵢ aᵢXᵢ > t] ≤ exp(-t²/(2σ²·∑ᵢ aᵢ²)).
theorem
corollary_1_7_lower_tail
{Ω : Type u_1}
[MeasurableSpace Ω]
{μ : MeasureTheory.Measure Ω}
[MeasureTheory.IsProbabilityMeasure μ]
{n : ℕ}
{X : Fin n → Ω → ℝ}
{σsq : ℝ}
(hX_sg : ∀ (i : Fin n), IsSubGaussian (X i) σsq μ)
(hX_indep : ProbabilityTheory.iIndepFun X μ)
(hX_meas : ∀ (i : Fin n), Measurable (X i))
(a : Fin n → ℝ)
(t : ℝ)
(ht : 0 < t)
:
Corollary 1.7 — Lower tail bound for linear combinations.
For independent sub-Gaussian random variables Xᵢ ~ subG(σ²) and any a : Fin n → ℝ,
P[∑ᵢ aᵢXᵢ < -t] ≤ exp(-t²/(2σ²·∑ᵢ aᵢ²)).
theorem
corollary_1_7_linear_combination_tail
{Ω : Type u_1}
[MeasurableSpace Ω]
{μ : MeasureTheory.Measure Ω}
[MeasureTheory.IsProbabilityMeasure μ]
{n : ℕ}
{X : Fin n → Ω → ℝ}
{σsq : ℝ}
(hX_sg : ∀ (i : Fin n), IsSubGaussian (X i) σsq μ)
(hX_indep : ProbabilityTheory.iIndepFun X μ)
(hX_meas : ∀ (i : Fin n), Measurable (X i))
(a : Fin n → ℝ)
(t : ℝ)
(ht : 0 < t)
:
Corollary 1.7 — Combined tail bound for linear combinations.
For independent sub-Gaussian random variables Xᵢ ~ subG(σ²), any a : Fin n → ℝ,
and t > 0, both tails satisfy exp(-t²/(2σ²·‖a‖₂²)).