taylorAux_hasDerivAt
theoremverifiedimported Lean
Derivative of a Taylor auxiliary function. Let $h \colon \bbr \to \bbr$ be the function $h(x) = x^2 + x + \log(1 - x)$. Then at every point $u < 1$, the function $h$ is differentiable with derivative \[ h'(u) = \frac{u(1 - 2u)}{1 - u}. \]
Statement
set_option relaxedAutoImplicit false in set_option autoImplicit false in open MeasureTheory ProbabilityTheory Real NNReal Matrix Finset in variable {d k : ℕ} in lemma taylorAux_hasDerivAt (u : ℝ) (hu : u < 1) : HasDerivAt (fun x : ℝ => x ^ 2 + x + Real.log (1 - x)) (u * (1 - 2 * u) / (1 - u)) u
- Source
- TCSlib, `taylorAux_hasDerivAt`
- Lean proof
- By Ganesh Sankar, from github.com/Shilun-Allan-Li/tcslib at
c8a02591b1e5(Copyright (c) 2026 Ganesh Sankar. All rights reserved, Apache-2.0). Compiled here against leanprover/lean4:v4.33.1, Mathlib v4.33.1; written for leanprover/lean4:v4.25.0.taylorAux_hasDerivAt← taylorAux_hasDerivAt (TCSlib/LearningTheory/JohnsonLindenstrauss/ChiSquaredMGF.lean:92): adapted
What was changed in
taylorAux_hasDerivAt--- TCSlib/LearningTheory/JohnsonLindenstrauss/ChiSquaredMGF.lean:92 (taylorAux_hasDerivAt) +++ as compiled here @@ -1,4 +1,6 @@ -/-- For `u < 1`, the auxiliary function `h(x) = x² + x + log(1 − x)` is differentiable at -`u` with derivative `u(1 − 2u)/(1 − u)`. -/ -private lemma taylorAux_hasDerivAt (u : ℝ) (hu : u < 1) : +set_option relaxedAutoImplicit false in +set_option autoImplicit false in +open MeasureTheory ProbabilityTheory Real NNReal Matrix Finset in +variable {d k : ℕ} in +lemma taylorAux_hasDerivAt (u : ℝ) (hu : u < 1) : HasDerivAt (fun x : ℝ => x ^ 2 + x + Real.log (1 - x)) @@ -16,3 +18,3 @@ -- Goal: 2u + 1 + (-1)/(1-u) = u(1-2u)/(1-u) - convert h5 using 1 + convert h5 using 1 <;> try rfl field_simp - Verified
- 07 Oct 2026
- Axioms
Classical.choiceQuot.soundpropext- Used by 1
- neg_log_one_sub_two_mul_le_two_sqTaylor bound for the centered chi-squared log-MGF. For every real number $s$ with $\abs{s} \le 1/4$, one has…
Lean source view module on GitHub
Choose a tactic — a marked line — to see the goals Lean had before and after it, as in Lean’s infoview. Step with ← → or j k.
set_option relaxedAutoImplicit false inset_option autoImplicit false inopen MeasureTheory ProbabilityTheory Real NNReal Matrix Finset invariable {d k : ℕ} in/-- For `u < 1`, the auxiliary function `h(x) = x² + x + log(1 − x)` is differentiable at `u` with derivative `u(1 − 2u)/(1 − u)`. -/lemma taylorAux_hasDerivAt (u : ℝ) (hu : u < 1) : HasDerivAt (fun x : ℝ => x ^ 2 + x + Real.log (1 - x)) (u * (1 - 2 * u) / (1 - u)) u := by have hne : (1 - u : ℝ) ≠ 0 := by linarith have h1 : HasDerivAt (fun x : ℝ => x ^ 2) (2 * u) u := by simpa using (hasDerivAt_pow 2 u) have h2 : HasDerivAt (fun x : ℝ => x) 1 u := hasDerivAt_id u have h3 : HasDerivAt (fun x : ℝ => 1 - x) (-1) u := (hasDerivAt_id u).const_sub 1 have h4 : HasDerivAt (fun x : ℝ => Real.log (1 - x)) (-1 / (1 - u)) u := h3.log hne have h5 := (h1.add h2).add h4 -- Sum derivative: 2u + 1 + (-1/(1-u)) -- Goal: 2u + 1 + (-1)/(1-u) = u(1-2u)/(1-u) convert h5 using 1 <;> try rfl field_simp ring