Real containment for interval arcsine and arccosine #
Mathlib's Real.arcsin_eq_arctan connects interior point enclosures to arcsine. Its
monotonicity theorem then bounds every real member of an interval by the endpoint images.
The cases at -1 and 1 use the existing enclosure of π/4. Arccosine containment follows
from Real.arccos_eq_pi_div_two_sub_arcsin.
@[simp]
Arcsine preserves exact zero independently of both accuracy parameters.
@[simp]
Arccosine at one is exactly zero independently of both accuracy parameters.
theorem
FloatLib.Numerics.Interval.Internal.containsReal_asinPointBounds?
{q : ℚ}
{terms precision : ℕ}
{K : Interval ℚ}
(h : asinPointBounds? q terms precision = some K)
:
ContainsReal some K (Real.arcsin ↑q)
A successful point enclosure contains the real arcsine, including at both endpoints.
theorem
FloatLib.Numerics.Interval.containsReal_asinBounds?
{I K : Interval ℚ}
{terms precision : ℕ}
(h : I.asinBounds? terms precision = some K)
{x : ℝ}
(hx : ContainsReal some I x)
:
ContainsReal some K (Real.arcsin x)
Successful arcsine bounds contain the image of every real member of the input interval.
theorem
FloatLib.Numerics.Interval.containsReal_acosBounds?
{I K : Interval ℚ}
{terms precision : ℕ}
(h : I.acosBounds? terms precision = some K)
{x : ℝ}
(hx : ContainsReal some I x)
:
ContainsReal some K (Real.arccos x)
Successful arccosine bounds contain the image of every real member of the input interval.