Local trials read distinct terminal cells when their leaves stay below the tested child
ProvedOAI.Problem310.WindowTrialAddresses.localTerminalAddress_injectiveSetting. Fix . The children of a vertex are labelled : child is the default child, and are the nondefault children, ordered as integers. A node is a finite sequence of children (the path from the root) of length , and a leaf is a sequence of children of length exactly ; denotes followed by the child . Bits are written (false) and (true). For and let
be the index of the cell containing in the -periodic grid of mesh .
Window data. Fix also . Call a sequence of children with an edge (it stands for the edge into its last vertex). A window datum consists of maps and from finite sequences of children to such that, for all edges :
- (i) and ;
- (ii) ;
- (iii) , so the window of consists of consecutive scale indices;
- (iv) if is a proper prefix of , then ;
- (v) if and with (for any sequences ), then ;
- (vi) if is a prefix of , then ;
- (vii) .
For a node , all child edges have windows of the same length .
Addresses. The selector table of a node and a nondefault child has resolution ; its cells are addressed by the pairs with . The terminal table of a leaf has resolution ; its cells are addressed by the pairs with . Addresses of different tables are different even when their cell indices coincide. For the center address of the selector table is .
Local trials. Fix a node . A local trial at is a pair with and ; its scale index runs through the window of the child edge . For the trial reads the selector table at the shifted point , i.e. at the address .
Theorem. Let be a window datum, a node, and . Let assign to every local trial at a leaf such that the child edge is a prefix of . Then the terminal addresses read by the trials,
are pairwise distinct, i.e. this map on is injective.
No numerical hypothesis is needed beyond the window axioms: they force the terminal resolution of every leaf below to refine the resolution of the child window. In the random model the leaf is reached by routing from and depends on the selector tables; applying the theorem for each selector outcome shows that distinct trials always read distinct terminal cells, so their terminal bits are independent.
Formalization Note Leaves are Leaf M d = List.Vector (Fin (M + 1)) d, the prefix condition is (selectorEdge P z.1).IsPrefix (leaf z).val with selectorEdge P i = P.val ++ [i.castSucc], the terminal resolution is terminalEndpoints W L = W.b L.val, and the address is terminalAddress (terminalEndpoints W) (leaf z) (x + t * (2 : ℝ)⁻¹ ^ localIndex W P z).
import Mathlib
import Definitions.Def_DyadicAvoidance
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_PeriodicBasic
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_GridSeparation
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_FiniteTableModel
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_OrderedRouting
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_RoutingPath
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_CenterRouteExposure
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_ExceptionalCenters
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_CenterFailureBound
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_FirstDefault
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_ConcreteLabelMeasures
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_CenterFirstDefault
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_RandomSetDensity
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_RoutedSetDensity
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_RoutedExposure
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_FirstDefaultSelection
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_FiniteLocalPredicate
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_WindowData
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_WindowTableModel
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_StableRoutingBridge
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_WindowExceptionalTransfer
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_TrialAddresses
import Definitions.Def_OAI_MeasureTheory_DyadicAvoidance_WindowTrialAddresses
open OAI OAI.Problem310 OAI.Problem310.WindowTrialAddresses
open FiniteTableModel TrialAddresses
variable {M d g r₀ : ℕ}theorem OAI.Problem310.WindowTrialAddresses.localTerminalAddress_injective (W : WindowData (M + 1) d g r₀)
(P : Node M d) (x t : ℝ) (ht : t ∈ Set.Icc (1 : ℝ) 2)
(leaf : LocalTrial W P → Leaf M d)
(hprefix : ∀ z, (selectorEdge P z.1).IsPrefix (leaf z).val) :
Function.Injective (fun z : LocalTrial W P =>
terminalAddress (terminalEndpoints W) (leaf z)
(x + t * ((2 : ℝ)⁻¹ ^ localIndex W P z))) := by sorry