potential
defimported Lean
The potential (sum of unnormalized weights) at time $t$ is \[ W_t \;=\; \sum_{i=1}^{N} w_t(i). \] The potential telescopes to relate the learner's cumulative loss to the best expert's loss.
Statement
set_option relaxedAutoImplicit false in set_option autoImplicit false in open Real Finset BigOperators in noncomputable def potential {N T : ℕ} (η : ℝ) (ℓ : LossSeq N T) (t : ℕ) : ℝ
- Defined in
- TCSlib, `potential`
- 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.potential← potential (TCSlib/LearningTheory/Hedge/Basic.lean:101): verbatim
- Verified
- 07 Oct 2026
- Axioms
- none listed
- Built on 2
- hedgeWeightThe unnormalized Hedge weight of expert $i$ at time $t$ with learning rate $\eta$ 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 15
- hedge_regret_bound_tightTight Hedge regret bound. Fix $N \ge 1$ experts and a horizon of $T$ rounds, and run the Hedge algorithm with a learning rate $\eta > 0$.…
- log_potential_step_tightTight per-step logarithmic potential bound for Hedge. Fix $N \ge 1$ experts and a learning rate $\eta > 0$, and let $\ell$ be a valid loss…
- hedgeDistThe Hedge distribution at round $t$ is the normalization of the weight vector: \[ p_t(i) \;=\; \frac{w_t(i)}{W_t}. \]
- potential_ratio_lePotential ratio bound for Hedge. Fix $N \ge 1$ experts, a horizon of $T$ rounds, a learning rate $\eta > 0$, and a valid loss sequence…
- potential_posPositivity of the Hedge potential. Fix a learning rate $\eta \in \bbr$, a loss sequence $\ell$ for $N \ge 1$ experts over $T$ rounds, and…
- prefixPotential_eq_potentialPrefix potential equals the induced Hedge potential. Let $G$ be a finite two-player zero-sum game with payoff matrix $A$ having $M$ rows…
- log_potential_telescopeTelescoping of the log potential. Let $\eta \in \bbr$ and let $\ell$ be a loss sequence for $N$ experts over $T$ rounds, with potential…
- op184_partially_adaptive_half_approximationFor every Pandora's problem instance with deadlines
inst : PandoraDeadlineInstance, there exists a valid partially adaptive policy… - additive_coloring_bound_by_degreeLet G = (V, E) be a finite simple graph, and let a : V → ℕ assign a positive integer to each vertex. Then there exists an additive…
- log_potential_stepOne-step logarithmic potential bound for Hedge. Fix $N \ge 1$ experts and $T$ rounds, a learning rate $\eta > 0$, and a valid loss…
- tarski_box_query_le_succ_halvingOne dimension costs a factor log₂ n: from a box-search algorithm in dimension k with q queries one gets one in dimension k+1 with at most…
- 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,…
- potential_zeroInitial value of the potential. Fix $N \ge 1$ experts, a horizon $T$, and a loss sequence $\ell$ over these experts and rounds, and let…
- potential_ge_best_expertPotential lower bound via the best expert. Fix $N \ge 1$ experts and $T$ rounds, a loss sequence $\ell$, and a learning rate $\eta > 0$.…
- hedge_regret_boundRegret bound for the Hedge algorithm. Fix $N \ge 1$ experts and $T$ rounds, and let $\ell$ be a valid loss sequence, so that…
Lean source view module on GitHub
set_option relaxedAutoImplicit false in set_option autoImplicit false in open Real Finset BigOperators in /-- The potential `W_t = ∑_i w_t(i)`, the sum of the unnormalized weights at round `t` [FS97, §2]; [CBL06, proof of Thm 2.2]. The potential is the object we track: its one-step upper bound gives Hedge's cumulative loss, while its final lower bound sees the best expert. -/ noncomputable def potential {N T : ℕ} (η : ℝ) (ℓ : LossSeq N T) (t : ℕ) : ℝ := ∑ i : Fin N, hedgeWeight η ℓ t i