Cslib.MachineLearning.PACLearning.error_map_eq_hypothesisError
theoremverifiedformalizationimported Lean
Under a realizable distribution P.map (x ↦ (x, c(x))), the general 0-1 error coincides with the binary hypothesisError P h c, where h and c are viewed as subsets of α via the characteristic function decide (· ∈ ·).
Statement
set_option autoImplicit true in set_option relaxedAutoImplicit true in open MeasureTheory Set in open scoped ENNReal NNReal in variable {α : Type*} [MeasurableSpace α] in open Classical in theorem Cslib.MachineLearning.PACLearning.error_map_eq_hypothesisError (P : Measure α) (h c : Set α) (hh : MeasurableSet h) (hc : MeasurableSet c) : error (P.map (fun x => (x, decide (x ∈ c)))) (fun x => decide (x ∈ h)) = hypothesisError P h c
- Formalizes
- CSLib, `Cslib.MachineLearning.PACLearning.error_map_eq_hypothesisError`
- Lean proof
- By Samuel Schlesinger, from github.com/leanprover/cslib at
3951377e5a3f(Copyright (c) 2026 Samuel Schlesinger. All rights reserved, Apache-2.0). Compiled here against leanprover/lean4:v4.33.1, Mathlib v4.33.1; written for leanprover/lean4:v4.33.0.Cslib.MachineLearning.PACLearning.error_map_eq_hypothesisError← Cslib.MachineLearning.PACLearning.error_map_eq_hypothesisError (Cslib/MachineLearning/PACLearning/Defs.lean:509): verbatim
- Verified
- 08 Oct 2026
- Axioms
Classical.choiceQuot.soundpropext- Built on 2
- Cslib.MachineLearning.PACLearning.errorThe *prediction error* (0-1 loss) of a hypothesis
hunder a joint distributionDonα × β, defined as the probability that the… - Cslib.MachineLearning.PACLearning.hypothesisErrorThe *symmetric-difference error* of a hypothesis
hwith respect to a target conceptc(both viewed as subsets ofα) under…
- Cslib.MachineLearning.PACLearning.errorThe *prediction error* (0-1 loss) of a hypothesis
Lean source view module on GitHub
Choose a tactic — a marked line — to see the goals Lean had before and after it, as in Lean’s infoview. Step with ← → or j k.
set_option autoImplicit true inset_option relaxedAutoImplicit true inopen MeasureTheory Set inopen scoped ENNReal NNReal invariable {α : Type*} [MeasurableSpace α] inopen Classical in/-- Under a realizable distribution `P.map (x ↦ (x, c(x)))`, the general 0-1 `error` coincides with the binary `hypothesisError P h c`, where `h` and `c` are viewed as subsets of `α` via the characteristic function `decide (· ∈ ·)`. -/theorem Cslib.MachineLearning.PACLearning.error_map_eq_hypothesisError (P : Measure α) (h c : Set α) (hh : MeasurableSet h) (hc : MeasurableSet c) : error (P.map (fun x => (x, decide (x ∈ c)))) (fun x => decide (x ∈ h)) = hypothesisError P h c := by simp only [error, hypothesisError] have hf : Measurable (fun x => (x, decide (x ∈ c))) := Measurable.prodMk measurable_id (measurable_to_bool (by convert hc using 1; ext x; simp [decide_eq_true_eq])) rw [Measure.map_apply_of_aemeasurable hf.aemeasurable] · congr 1; ext x by_cases hx : x ∈ h <;> simp_all [symmDiff_def] · convert (hh.prod (measurableSet_singleton false)).union (hh.compl.prod (measurableSet_singleton true)) using 1 ext ⟨x, b⟩; cases b <;> simp