Cslib.MachineLearning.PACLearning.ae_mem_versionSpace_of_realizable
theoremverifiedformalizationimported Lean
Under iid sampling from the realizable joint distribution induced by c ∈ C and a probability measure P on α, the target concept c lies in the version space almost surely.
Statement
set_option autoImplicit true in set_option relaxedAutoImplicit true in open MeasureTheory Set in open scoped ENNReal in variable {α : Type*} {β : Type*} in theorem Cslib.MachineLearning.PACLearning.ae_mem_versionSpace_of_realizable [MeasurableSpace α] [MeasurableSpace β] {C : ConceptClass α β} {c : α → β} (hc : c ∈ C) (hcm : Measurable c) (hG : MeasurableSet {p : α × β | p.2 = c p.1}) (P : Measure α) [IsProbabilityMeasure P] (m : ℕ) : ∀ᵐ S : LabeledSample α β m ∂(Measure.pi (fun _ : Fin m => P.map (fun x => (x, c x)))), c ∈ VersionSpace C S
- Formalizes
- CSLib, `Cslib.MachineLearning.PACLearning.ae_mem_versionSpace_of_realizable`
- Lean proof
- By Dhruv Gupta, from github.com/leanprover/cslib at
3951377e5a3f(Copyright (c) 2026 Dhruv Gupta. 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.ae_mem_versionSpace_of_realizable← Cslib.MachineLearning.PACLearning.ae_mem_versionSpace_of_realizable (Cslib/MachineLearning/PACLearning/VersionSpace.lean:263): verbatim
- Verified
- 08 Oct 2026
- Axioms
Classical.choiceQuot.soundpropext- Built on 4
- Cslib.MachineLearning.PACLearning.LabeledSampleA *labeled sample* of size
mover domainαwith label typeβis a finite sequence of(point, label)pairs. - Cslib.MachineLearning.PACLearning.VersionSpaceThe *version space* of a concept class
Cgiven a labeled sampleS: the set of concepts inCwhose labels agree withSon every… - Cslib.MachineLearning.PACLearning.ConceptClassA *concept class* over domain
αwith label typeβis a set of functionsα → β. For binary classification (β = Bool), this is… - Cslib.MachineLearning.PACLearning.pi_map_graph_eq_oneThe iid product of the realizable joint distribution assigns measure
1to the set of samples where every coordinate lies on the graph of…
- Cslib.MachineLearning.PACLearning.LabeledSampleA *labeled sample* of size
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 invariable {α : Type*} {β : Type*} in/-- Under iid sampling from the realizable joint distribution induced by `c ∈ C` and a probability measure `P` on `α`, the target concept `c` lies in the version space almost surely. -/theorem Cslib.MachineLearning.PACLearning.ae_mem_versionSpace_of_realizable [MeasurableSpace α] [MeasurableSpace β] {C : ConceptClass α β} {c : α → β} (hc : c ∈ C) (hcm : Measurable c) (hG : MeasurableSet {p : α × β | p.2 = c p.1}) (P : Measure α) [IsProbabilityMeasure P] (m : ℕ) : ∀ᵐ S : LabeledSample α β m ∂(Measure.pi (fun _ : Fin m => P.map (fun x => (x, c x)))), c ∈ VersionSpace C S := by have hφ : Measurable (fun x : α => (x, c x)) := by fun_prop have : IsProbabilityMeasure (P.map (fun x : α => (x, c x))) := Measure.isProbabilityMeasure_map hφ.aemeasurable rw [ae_iff] have hsub : {S : Fin m → α × β | ¬ c ∈ VersionSpace C S} ⊆ (Set.univ.pi (fun _ : Fin m => {p : α × β | p.2 = c p.1}))ᶜ := by intro S hS hcontra exact hS ⟨hc, by simp_all⟩ have hcompl : (Measure.pi (fun _ : Fin m => P.map (fun x : α => (x, c x)))) ((Set.univ.pi (fun _ : Fin m => {p : α × β | p.2 = c p.1}))ᶜ) = 0 := by rw [prob_compl_eq_one_sub (MeasurableSet.univ_pi fun _ => hG), pi_map_graph_eq_one hcm P hG, tsub_self] exact measure_mono_null hsub hcompl