Exact minimax characterization of finite-horizon k-server guarantees
ProvedKServer.finite_game_characterizationcompetitive-analysisdynamic-programmingk-server
Fix , an arbitrary metric space , an initial configuration , a finite request set , a horizon , an arbitrary real-valued prefix payoff , and a real threshold . Let denote the finite stopping-game recursion of KServer_finite_game. Then
The algorithm is deterministic and online, and must serve all histories in , including those outside the finite test set. The guarantee concerns every permitted stopping prefix. No regularity, sign, monotonicity, or offline-optimality assumption is imposed on .
This identifies the exact scalar threshold for existence of a finite-horizon online strategy. It justifies using the finite recursion both to construct algorithms and to rule out proposed competitive bounds on a fixed finite game.
Preamble
import Definitions.Def_KServer_finite_game open KServer
Formal statement
theorem KServer.finite_game_characterization (k : ℕ) (hk : 1 ≤ k) (M : Type) [MetricSpace M]
(C₀ : Config k M) (P : Finset M) (b : List M → ℝ) (n : ℕ) (a : ℝ) :
finiteGameValue hk P b n [] C₀ ≤ a ↔
∃ A : OnlineAlgorithm k M, A.conf [] = C₀ ∧
∀ σ : List M, σ.length ≤ n → (∀ r ∈ σ, r ∈ P) →
A.cost σ ≤ b σ + a := by sorrySource
Original backward-induction formulation for KServer.finite_horizon_uniform_bound (10ab1333-61a5-4d90-ac6b-b692e2316d1e). The game recursion is specified in this definition, and its exact algorithmic interpretation is proved in KServer.finite_game_characterization. Source of the underlying open conjecture: Koutsoupias, The k-server problem (2009), Conjecture 1, preprint p. 2, https://www.cs.ox.ac.uk/people/elias.koutsoupias/Personal/Papers/paper-kou09.pdf. This recurrence and characterization are an original formal development, not a claim that the conjecture is proved in that reference.