Recursive Definitions as Selected Fixed Points
Abstract
Recursive definitions are fixed points with explicit extremal selections.
Theorem 1.1 (A recursive equation is a fixed-point equation).
Proof. Machine-checked in Lean as D5/S1/Dynamics/RecursiveDefinition.is_recursive_definition_iff_fixed_point (✓ std3). ∎
Source. Repository-derived.
Commentary.
For an arbitrary endomorphism and candidate value, the equation f(x) = x is equivalent to membership in Function.fixedPoints f.
Theorem 1.2 (Distinct extremal fixed points make the selection observable).
Proof. Machine-checked in Lean as D5/S1/Dynamics/RecursiveDefinition.extremal_selection_distinguishes_fixed_points (✓ std3). ∎
Source. Repository-derived.
Commentary.
The selector is explicit data with least and greatest cases. For a monotone endomorphism of a complete lattice, if its least and greatest fixed points differ, both selected values satisfy the fixed-point equation and the two selected values are unequal.
Theorem 1.3 (Uniqueness identifies the least and greatest fixed points).
Proof. Machine-checked in Lean as D5/S1/Dynamics/RecursiveDefinition.unique_fixed_point_implies_lfp_eq_gfp (✓ std3). ∎
Source. Repository-derived.
Commentary.
For a monotone endomorphism of a complete lattice, the existence of exactly one value satisfying f(x) = x implies that the least and greatest fixed points coincide.
References
- Truth anchor:
D5/S1/Dynamics/RecursiveDefinition.extremal_selection_distinguishes_fixed_points - Truth anchor:
D5/S1/Dynamics/RecursiveDefinition.is_recursive_definition_iff_fixed_point - Truth anchor:
D5/S1/Dynamics/RecursiveDefinition.unique_fixed_point_implies_lfp_eq_gfp - Dependency: D5/S1/Dynamics/KnasterTarski