Bridgeland Stability Conditions

14.16. Deformation: PhasePerturbation🔗

14.16.1. wPhaseOf_perturbation_generic🔗

Generic phase perturbation bound. If w = m \cdot e^{i\pi\phi} \cdot (1 + u) with m > 0, \phi \in (\alpha - 1/2, \alpha + 1/2), and \|u\| < \sin(\pi\varepsilon) with 0 < \varepsilon \le 1/2, then |\operatorname{wPhaseOf}(w, \alpha) - \phi| < \varepsilon.

Proof: Split \arg of the product using \arg(z_1 z_2) = \arg(z_1) + \arg(z_2) (valid when the sum stays in (-\pi, \pi]). Compute \arg(m \cdot e^{i\pi(\phi - \alpha)}) = \pi(\phi - \alpha), and bound |\arg(1 + u)| < \pi\varepsilon from the pure complex analysis lemma abs_arg_one_add_lt.

🔗theorem
CategoryTheory.Triangulated.wPhaseOf_perturbation_generic {m φ α ε : } {u : } (hm : 0 < m) ( : φ Set.Ioo (α - 1 / 2) (α + 1 / 2)) ( : 0 < ε) (hε2 : ε 1 / 2) (hu : u < Real.sin (Real.pi * ε)) : |CategoryTheory.Triangulated.wPhaseOf (m * Complex.exp ((Real.pi * φ) * Complex.I) * (1 + u)) α - φ| < ε
CategoryTheory.Triangulated.wPhaseOf_perturbation_generic {m φ α ε : } {u : } (hm : 0 < m) ( : φ Set.Ioo (α - 1 / 2) (α + 1 / 2)) ( : 0 < ε) (hε2 : ε 1 / 2) (hu : u < Real.sin (Real.pi * ε)) : |CategoryTheory.Triangulated.wPhaseOf (m * Complex.exp ((Real.pi * φ) * Complex.I) * (1 + u)) α - φ| < ε

Something wrong, better idea? Suggest a change