OAI/Probability/MatroidSecretary/Labels/LabeledTransport.lean

GitHub source file

21 declarations, 0 contain sorry, 0 declared as axioms

  1. def OAI.MatroidProphet.LabeledTransport.optimum line 14

    Source documentation

    The same offline maximum on an arbitrary finite label type.

    noncomputable def optimum (M : Matroid α) (w : α → ℝ) : ℝ := by
    …
  2. theorem OAI.MatroidProphet.LabeledTransport.sum_le_optimum line 19
  3. theorem OAI.MatroidProphet.LabeledTransport.optimum_nonneg line 27
  4. theorem OAI.MatroidProphet.LabeledTransport.mapEquiv_ground line 32

    Source documentation

    Relabeling preserves the full ground set; it does not discard loops.

    theorem mapEquiv_ground (M : Matroid α) (hE : M.E = Set.univ)
        (e : α ≃ Fin n) : (M.mapEquiv e).E = Set.univ := by
    …
  5. theorem OAI.MatroidProphet.LabeledTransport.optimum_eq_indexed line 37

    Source documentation

    The all-independent-sets benchmark is exactly unchanged by label encoding.

    theorem optimum_eq_indexed (M : Matroid α) (e : α ≃ Fin n) (w : α → ℝ) :
        optimum M w = MatroidProphet.optimum (M.mapEquiv e) (w ∘ e.symm) := by
    …

    Mentions: OAI.MatroidProphet.LabeledTransport.optimum, OAI.MatroidProphet.optimum

  6. abbrev OAI.MatroidProphet.LabeledTransport.History line 59
  7. abbrev OAI.MatroidProphet.LabeledTransport.ArrivalOrder line 62

    Source documentation

    A physical arrival order visits every label once, indexed by arrival time.

    abbrev ArrivalOrder (α : Type*) (n : ℕ) := Fin n ≃ α
  8. instance OAI.MatroidProphet.LabeledTransport.arrivalOrderMeasurableSpace line 64
  9. instance OAI.MatroidProphet.LabeledTransport.arrivalOrderMeasurableSingletonClass line 66
  10. structure OAI.MatroidProphet.LabeledTransport.OnlineRule line 72

    Source documentation

    A rule on physical labels receives only the same sample and history data.

    structure OnlineRule (α : Type*) [MeasurableSpace α] (n bits : ℕ) where
    …
  11. def OAI.MatroidProphet.LabeledTransport.encodeHistory line 77
  12. def OAI.MatroidProphet.LabeledTransport.decodeRule line 81

    Source documentation

    Decode the indexed rule without adding any future-information input.

    noncomputable def decodeRule (e : α ≃ Fin n) (A : MatroidProphet.OnlineRule n bits) :
        OnlineRule α n bits where
    …

    Mentions: OAI.MatroidProphet.OnlineRule, OAI.MatroidProphet.LabeledTransport.OnlineRule

  13. def OAI.MatroidProphet.LabeledTransport.history line 102
  14. def OAI.MatroidProphet.LabeledTransport.decisionAt line 105
  15. def OAI.MatroidProphet.LabeledTransport.acceptedThrough line 108
  16. def OAI.MatroidProphet.LabeledTransport.reward line 113
  17. theorem OAI.MatroidProphet.LabeledTransport.encode_history line 118
  18. theorem OAI.MatroidProphet.LabeledTransport.decode_decision line 124
  19. theorem OAI.MatroidProphet.LabeledTransport.decode_acceptedThrough line 130
  20. theorem OAI.MatroidProphet.LabeledTransport.decode_reward line 142
  21. theorem OAI.MatroidProphet.LabeledTransport.decode_feasible line 149

    Source documentation

    Prefix feasibility transports back to the original labels, including loops.

    theorem decode_feasible (M : Matroid α) (e : α ≃ Fin n)
        (A : MatroidProphet.OnlineRule n bits) (hA : Feasible (M.mapEquiv e) A)
        (r : Seed bits) (s w : α → ℝ) (π : ArrivalOrder α n) (t : ℕ)
        (hs : ∀ a, 0 ≤ s a) (hw : ∀ a, 0 ≤ w a) :
        M.Indep (acceptedThrough (decodeRule e A) r s w π t : Set α) := by
    …

    Mentions: OAI.MatroidProphet.OnlineRule, OAI.MatroidProphet.Feasible, OAI.MatroidProphet.Seed, OAI.MatroidProphet.LabeledTransport.ArrivalOrder, OAI.MatroidProphet.LabeledTransport.acceptedThrough, OAI.MatroidProphet.LabeledTransport.decodeRule