AFTD Auto-Formalizing Theoretical Domains

hedgeCumLoss

defimported Lean

The cumulative loss of the Hedge algorithm over all $T$ rounds is \[ \widehat{L}_T \;=\; \sum_{t=0}^{T-1} \widehat{\ell}_t. \]

Statement

set_option relaxedAutoImplicit false in
set_option autoImplicit false in
open Real Finset BigOperators in
noncomputable def hedgeCumLoss {N T : ℕ} (η : ℝ) (ℓ : LossSeq N T) : ℝ
Defined in
TCSlib, `hedgeCumLoss`
Lean proof
By Karim Abdel Sadek, Mark Bedaywi, from github.com/Shilun-Allan-Li/tcslib at c8a02591b1e5 (Copyright (c) 2026 Karim Abdel Sadek and Mark Bedaywi. All rights reserved, Apache-2.0). Compiled here against leanprover/lean4:v4.33.1, Mathlib v4.33.1; written for leanprover/lean4:v4.25.0.
Verified
07 Oct 2026
Axioms
none listed
Built on 2
  • hedgeLossThe expected loss of the learner at round $t$ under the Hedge distribution is…
  • LossSeqA loss sequence $\ell$ for $N$ experts over $T$ rounds is a function $\ell : \mathrm{Fin}\,T \to \mathrm{Fin}\,N \to \mathbb{R}$,…
Used by 3
  • hedge_constructionHedge yields an approximate minimax pair. Let $G$ be a finite two-player zero-sum game with $M \ge 2$ row actions and $N \ge 1$ column…
  • regretThe regret of the Hedge algorithm is the excess cumulative loss over the best expert in hindsight:…
  • regret_le_of_log_potential_stepRegret from a per-step log-potential bound. Let $N \ge 1$, let $\eta > 0$, let $\ell$ be a loss sequence for $N$ experts over $T$ rounds,…

Lean source view module on GitHub

set_option relaxedAutoImplicit false in
set_option autoImplicit false in
open Real Finset BigOperators in
/-- The cumulative expected loss `∑_{t < T} ∑_i p_t(i) ℓ_t(i)` of Hedge over the `T` rounds [FS97, §2]; [CBL06, §2.1]. -/
noncomputable def hedgeCumLoss {N T : ℕ} (η : ℝ) (ℓ : LossSeq N T) : ℝ :=
  ∑ t : Fin T, hedgeLoss η ℓ t
See it in the graphDiscuss this resultChallenge it