AFTD Auto-Formalizing Theoretical Domains

nae_sine_contraction

theoremverified

For 0 ≤ t ≤ π/2, (√8/3) sin²(t) ≤ sin(15t/16).

Statement

theorem nae_sine_contraction (t : ℝ) (ht0 : 0 ≤ t) (ht : t ≤ Real.pi / 2) :
    Real.sqrt 8 / 3 * Real.sin t ^ 2 ≤ Real.sin ((15 / 16 : ℝ) * t)
Source
arXiv:1211.0721, Section 3.3, p-computation rule analysis
Verified
14 Nov 2026
Axioms
Classical.choiceQuot.soundpropext
Built on 3
Used by 1

The proof, explained

The proof divides the interval [0, π/2] into two cases at t = 7π/16. For t ≤ 7π/16, the bound sin²(t) ≤ 63/64 implies (√8/3) sin(t) ≤ 15/16, from which (√8/3) sin²(t) ≤ (15/16) sin(t) ≤ sin(15t/16) via a sine scaling inequality. For t ≥ 7π/16, the monotonicity of sine ensures sin(15t/16) ≥ sin(2π/5), and chaining (√8/3) sin²(t) ≤ √8/3 ≤ sin(2π/5) completes the estimate.

  1. The values sin(t) and √8/3 are non-negative.

    Since 0 ≤ t ≤ π/2 < π, the sine function is non-negative on [0, π] (by Real.sin_nonneg_of_nonneg_of_le_pi, using π > 0 from Real.pi_pos). The constant √8/3 is non-negative because 8 and 3 are positive and the square root of a positive real is positive.

    lines 3–5

    uses Real.sin_nonneg_of_nonneg_of_le_pi, Real.pi_pos

      have hs0 : 0 ≤ Real.sin t :=
        Real.sin_nonneg_of_nonneg_of_le_pi ht0 (by linarith [Real.pi_pos])
      have hc0 : 0 ≤ Real.sqrt 8 / 3 := by positivity
  2. The squared coefficient satisfies (√8/3)² = 8/9.

    Distributing the exponent over the quotient via div_pow gives (√8)² / 3², which simplifies to 8/9 using Real.sq_sqrt on 8 ≥ 0 and numerical evaluation.

    lines 6–8

    uses div_pow, Real.sq_sqrt

      have hc2 : (Real.sqrt 8 / 3) ^ 2 = (8 / 9 : ℝ) := by
        rw [div_pow, Real.sq_sqrt (by norm_num : (0 : ℝ) ≤ 8)]
        norm_num
  3. It suffices to prove the inequality in two cases: t ≤ 7π/16 and t > 7π/16.

    We perform a case split depending on whether t ≤ 7π/16.

    line 9

      by_cases hsmall : t ≤ 7 * Real.pi / 16
  4. When t ≤ 7π/16, the coefficient bound (√8/3) sin(t) ≤ 15/16 holds.

    By nae_low_sine_square, sin²(t) ≤ 63/64. Multiplying on the left by (√8/3)² ≥ 0 (via sq_nonneg and mul_le_mul_of_nonneg_left) and substituting (√8/3)² = 8/9 gives ((√8/3) sin(t))² ≤ (8/9)(63/64) = 7/8 = 224/256. Since 224/256 < (15/16)² = 225/256, non-linear arithmetic using ((√8/3) sin(t) - 15/16)² ≥ 0 deduces (√8/3) sin(t) ≤ 15/16.

    lines 10–14

    uses nae_low_sine_square, mul_le_mul_of_nonneg_left, sq_nonneg

      · have hsq := nae_low_sine_square t ht0 hsmall
        have hprod := mul_le_mul_of_nonneg_left hsq (sq_nonneg (Real.sqrt 8 / 3))
        rw [hc2] at hprod
        have hcoeff : Real.sqrt 8 / 3 * Real.sin t ≤ (15 / 16 : ℝ) := by
          nlinarith [sq_nonneg (Real.sqrt 8 / 3 * Real.sin t - (15 / 16 : ℝ))]
  5. For t ≤ 7π/16, the main inequality (√8/3) sin²(t) ≤ sin(15t/16) holds.

    Multiplying (√8/3) sin(t) ≤ 15/16 by sin(t) ≥ 0 on the right (via mul_le_mul_of_nonneg_right) gives (√8/3) sin²(t) ≤ (15/16) sin(t). The scaling inequality nae_sin_scale provides (15/16) sin(t) ≤ sin(15t/16) for 0 ≤ t ≤ π (where t ≤ π follows from t ≤ π/2 and Real.pi_pos). Combining these two inequalities concludes this case.

    lines 15–17

    uses mul_le_mul_of_nonneg_right, nae_sin_scale, Real.pi_pos

        have hmul := mul_le_mul_of_nonneg_right hcoeff hs0
        have hscale := nae_sin_scale t ht0 (by linarith [Real.pi_pos])
        nlinarith
  6. When t > 7π/16, we have 7π/16 ≤ t and sin(2π/5) ≤ sin(15t/16).

    From ¬(t ≤ 7π/16) we obtain 7π/16 ≤ t via le_of_not_ge. Then (15/16)t ≥ (15/16)(7π/16) = 105π/256 ≥ 2π/5, and (15/16)t ≤ (15/16)(π/2) ≤ π/2. Because both angles lie in [-π/2, π/2] and 2π/5 ≤ 15t/16, the monotonicity of sine on this interval (Real.sin_le_sin_of_le_of_le_pi_div_two, using Real.pi_pos) implies sin(2π/5) ≤ sin(15t/16).

    lines 18–23

    uses le_of_not_ge, Real.sin_le_sin_of_le_of_le_pi_div_two, Real.pi_pos

      · have hlarge : 7 * Real.pi / 16 ≤ t := le_of_not_ge hsmall
        have hmono : Real.sin (2 * Real.pi / 5) ≤ Real.sin ((15 / 16 : ℝ) * t) := by
          apply Real.sin_le_sin_of_le_of_le_pi_div_two
          · linarith [Real.pi_pos]
          · nlinarith [Real.pi_pos]
          · nlinarith [Real.pi_pos]
  7. For t > 7π/16, the main inequality (√8/3) sin²(t) ≤ sin(15t/16) holds.

    Since sin²(t) ≤ 1 (by Real.sin_sq_le_one), multiplying by √8/3 ≥ 0 on the left (via mul_le_mul_of_nonneg_left) gives (√8/3) sin²(t) ≤ √8/3. Using the constant bound √8/3 ≤ sin(2π/5) from nae_high_sine_fixed and the lower bound sin(2π/5) ≤ sin(15t/16), chaining the inequalities finishes the proof.

    lines 24–26

    uses Real.sin_sq_le_one, mul_le_mul_of_nonneg_left, nae_high_sine_fixed

        have hsq := Real.sin_sq_le_one t
        have hmul := mul_le_mul_of_nonneg_left hsq hc0
        nlinarith [nae_high_sine_fixed]
Real.sin_nonneg_of_nonneg_of_le_pi
The sine of any real number in [0, π] is non-negative.
Real.pi_pos
The mathematical constant π is strictly positive.
div_pow
The power of a quotient equals the quotient of the powers: (a / b)^n = a^n / b^n.
Real.sq_sqrt
For any non-negative real number x, (√x)^2 = x.
nae_low_sine_square
For 0 ≤ t ≤ 7π/16, sin²(t) ≤ 63/64.
mul_le_mul_of_nonneg_left
Multiplying both sides of an inequality by a non-negative quantity on the left preserves the inequality.
sq_nonneg
The square of any element in an ordered ring is non-negative: a² ≥ 0.
mul_le_mul_of_nonneg_right
Multiplying both sides of an inequality by a non-negative quantity on the right preserves the inequality.
nae_sin_scale
For 0 ≤ t ≤ π, (15/16) sin(t) ≤ sin(15t/16).
le_of_not_ge
In a linear order, if a is not less than or equal to b, then b ≤ a.
Real.sin_le_sin_of_le_of_le_pi_div_two
The sine function is monotonically non-decreasing on the interval [-π/2, π/2].
Real.sin_sq_le_one
For any real number x, sin²(x) ≤ 1.
nae_high_sine_fixed
The constant inequality √8/3 ≤ sin(2π/5) holds.

Written by a model (gemini-3.8-flash) from the Lean proof and the goal states Lean reported at each step. Lean verified the proof; this explanation is a reading aid and is not itself machine-checked.

Lean source view module on GitHub

theorem nae_sine_contraction (t : ℝ) (ht0 : 0 ≤ t) (ht : t ≤ Real.pi / 2) :    Real.sqrt 8 / 3 * Real.sin t ^ 2 ≤ Real.sin ((15 / 16 : ℝ) * t) := by  have hs0 : 0 ≤ Real.sin t :=    Real.sin_nonneg_of_nonneg_of_le_pi ht0 (by linarith [Real.pi_pos])  have hc0 : 0 ≤ Real.sqrt 8 / 3 := by positivity  have hc2 : (Real.sqrt 8 / 3) ^ 2 = (8 / 9 : ℝ) := by    rw [div_pow, Real.sq_sqrt (by norm_num : (0 : ℝ) ≤ 8)]    norm_num  by_cases hsmall : t ≤ 7 * Real.pi / 16  · have hsq := nae_low_sine_square t ht0 hsmall    have hprod := mul_le_mul_of_nonneg_left hsq (sq_nonneg (Real.sqrt 8 / 3))    rw [hc2] at hprod    have hcoeff : Real.sqrt 8 / 3 * Real.sin t ≤ (15 / 16 : ℝ) := by      nlinarith [sq_nonneg (Real.sqrt 8 / 3 * Real.sin t - (15 / 16 : ℝ))]    have hmul := mul_le_mul_of_nonneg_right hcoeff hs0    have hscale := nae_sin_scale t ht0 (by linarith [Real.pi_pos])    nlinarith  · have hlarge : 7 * Real.pi / 16 ≤ t := le_of_not_ge hsmall    have hmono : Real.sin (2 * Real.pi / 5) ≤ Real.sin ((15 / 16 : ℝ) * t) := by      apply Real.sin_le_sin_of_le_of_le_pi_div_two      · linarith [Real.pi_pos]      · nlinarith [Real.pi_pos]      · nlinarith [Real.pi_pos]    have hsq := Real.sin_sq_le_one t    have hmul := mul_le_mul_of_nonneg_left hsq hc0    nlinarith [nae_high_sine_fixed]
See it in the graphDiscuss this resultChallenge it