Samplinglib
Lean gate not recorded for this source state main · 9bb7a7394ab0

ActualPrescribedWeightedDensity: mathematical reading route

Read the statements and derivations in source order. Every result has its own optional Lean statement and proof. Source assumptions, library reuse and unproved boundaries are kept explicit.

  1. Actual PBPS weighted density under finite path reversal
Samplinglib mathematical lesson

Actual PBPS weighted density under finite path reversal

AutoSamplingTheory.ExampleCases.ProximalBPS.ActualPrescribedWeightedDensity.actual_prescribed_weighted_density_balance · theorem · Teaching coverage

Read this first

Mathematical statement

Let E be a finite-dimensional real inner-product space equipped with its Borel sigma algebra; zero dimension is allowed. Let alpha,beta≥0, eta be real, V:E→R. Assume exactly alpha>0, alpha≤beta, V is C2, alpha||v||^2≤D2V(x)[v,v]≤beta||v||^2 for every x,v, eta>0 and beta*eta≤1. All conclusions hold for every y,r in E, initial phase z0=(x0,p0), integer n≥0 and finite T≥0. Set c=y-eta*grad V(r). For z=(x,p), define Phi_u(z)=(c+cos(u)(x-c)+sqrt(eta)sin(u)p, -sin(u)/sqrt(eta)(x-c)+cos(u)p). Define h_r(x)=grad V(x)-grad V(r), S_r(x,p)=(x,p-[2<p,h_r(x)>/||h_r(x)||^2]h_r(x)) with real division totalized by 0/0=0, lambda_r(x,p)=sqrt(eta)max(0,<p,h_r(x)>), and Lambda_z(u)=integral from0 to u of lambda_r(Phi_s(z)) ds for u≥0, using Lebesgue measure. A tagged record is either live(a,z), with a≥0, or stopped; Hist_n is the type of all n+1 records. Let Delta_n(T) be the strict simplex 0<t1<...<tn<T. Write a0=0, ai=ti (1≤i≤n), a_(n+1)=T. Given t in Delta_n(T), define z_i=S_r(Phi_(a_i-a_(i-1))(z_(i-1))) for1≤i≤n and K_n(t)_j=live(a_j,z_j). The literal Lean nonnegative conversions preserve these nonnegative times. Define qReal_n(t)=product_(i=1)^n [lambda_r(Phi_(a_i-a_(i-1))(z_(i-1))) exp(-Lambda_(z_(i-1))(a_i-a_(i-1)))], q_n=ofReal(qReal_n). For arbitrary histories let readTime(live(a,z))=a and readTime(stopped)=0; d_n(T,h)=max(T-readTime(h_n),0). Put W(live(a,z),u)=ofReal(exp(-Lambda_z(u))) and W(stopped,u)=0. Define rho_n(T,t)=q_n(t) W(K_n(t)_n,d_n(T,K_n(t))). Here ofReal(x) is max(x,0) embedded into extended nonnegative reals; no probability law or normalization is assumed in this deterministic identity. Define Path_T as all phase-valued functions on[0,T]. The total interpolation Psi_n(h)(s) starts at z0 and scans records j=0,...,n in order: a live(a,z) record with a≤s replaces the accumulator by Phi_(max(s-a,0))(z); otherwise it is unchanged. Set Gamma_n(t)=Psi_n(K_n(t)). Extend this particular prescribed path to all real times by gamma(s)=Gamma_n(t)(min(max(s,0),T)). This extension agrees with Gamma on[0,T]; no topology on an entire function space is asserted. Define momentum reversal R(x,p)=(x,-p), target energy U_y(x,p)=V(x)+||x-y||^2/(2 eta)+||p||^2/2 and unnormalized Gibbs weight w_y(z)=exp(-U_y(z)). For any actual initial state z and strict event tuple t, let P_(y,r,z,n,T)(t)=product_(i=1)^n lambda_r(Phi_(a_i-a_(i-1))(z_(i-1))), I_(y,r,z,n,T)(t)=integral_0^T lambda_r(gamma(s)) ds and dReal_(y,r,z,n,T)(t)=exp(-I_(y,r,z,n,T)(t))*P_(y,r,z,n,T)(t). The incoming-rate product excludes the terminal arrival. The original ENNReal density rho_n(T,t) remains q_n times the terminal live survival factor defined above; it is not redefined as ofReal(dReal). For each n≥0 and T≥0 there exists a single reversal map rho:Prefix_n(T)→Prefix_n(T), with (rho(t))_i=T-t_(n+1-i) in one-based coordinates. For every y,r,z and strict tuple t, let A=Gamma_(y,r,z,n,T,t)(T). Then both identities hold: w_y(z) dReal_(y,r,z,n,T)(t) = w_y(R(A)) dReal_(y,r,R(A),n,T)(rho(t)), ofReal(w_y(z)) rho_(y,r,z,n,T)(t) = ofReal(w_y(R(A))) rho_(y,r,R(A),n,T)(rho(t)). The reversal map and all internal identities are produced, not assumed. The six original analytic assumptions are the only analytic premises. There is no positivity/nonzero-rate premise and no cancellation of event rates. The conclusions include n=0 with empty product1, T=0, zero-dimensional phase space, zero normals and zero rates; for T=0 and n>0 the strict event subtype is empty. This is unnormalized pointwise weighted density matching in Appendix A.1, not a weighted integral/sector identity, normalized invariant probability, Markov property, hypocoercivity, implementation error, actual expected cost, paper main result or PBPS/SPHMC composition.

\[\operatorname{ofReal}(w(z))\rho_{z,t}=\operatorname{ofReal}(w(RA))\rho_{RA,\rho(t)}.\]
Lean statement · actual_prescribed_weighted_density_balance

Let E be a finite-dimensional real inner-product space equipped with its Borel sigma algebra; zero dimension is allowed. Let alpha,beta>=0, eta be real, V:E->R. Assume exactly alpha>0, alpha<=beta, V is C2, alpha||v||^2<=D2V(x)[v,v]<=beta||v||^2 for every x,v, eta>0 and beta*eta<=1. All conclusions hold for every y,r in E, initial phase z0=(x0,p0), integer n>=0 and finite T>=0. Set c=y-eta*grad V(r). For z=(x,p), define Phi_u(z)=(c+cos(u)(x-c)+sqrt(eta)sin(u)p, -sin(u)/sqrt(eta)(x-c)+cos(u)p). Define h_r(x)=grad V(x)-grad V(r), S_r(x,p)=(x,p-[2<p,h_r(x)>/||h_r(x)||^2]h_r(x)) with real division totalized by 0/0=0, lambda_r(x,p)=sqrt(eta)max(0,<p,h_r(x)>), and Lambda_z(u)=integral from0 to u of lambda_r(Phi_s(z)) ds for u>=0, using Lebesgue measure. A tagged record is either live(a,z), with a>=0, or stopped; Hist_n is the type of all n+1 records. Let Delta_n(T) be the strict simplex 0<t1<...<tn<T. Write a0=0, ai=ti (1<=i<=n), a_(n+1)=T. Given t in Delta_n(T), define z_i=S_r(Phi_(a_i-a_(i-1))(z_(i-1))) for1<=i<=n and K_n(t)_j=live(a_j,z_j). The literal Lean nonnegative conversions preserve these nonnegative times. Define qReal_n(t)=product_(i=1)^n [lambda_r(Phi_(a_i-a_(i-1))(z_(i-1))) exp(-Lambda_(z_(i-1))(a_i-a_(i-1)))], q_n=ofReal(qReal_n). For arbitrary histories let readTime(live(a,z))=a and readTime(stopped)=0; d_n(T,h)=max(T-readTime(h_n),0). Put W(live(a,z),u)=ofReal(exp(-Lambda_z(u))) and W(stopped,u)=0. Define rho_n(T,t)=q_n(t) W(K_n(t)_n,d_n(T,K_n(t))). Here ofReal(x) is max(x,0) embedded into extended nonnegative reals; no probability law or normalization is assumed in this deterministic identity. Define Path_T as all phase-valued functions on[0,T]. The total interpolation Psi_n(h)(s) starts at z0 and scans records j=0,...,n in order: a live(a,z) record with a<=s replaces the accumulator by Phi_(max(s-a,0))(z); otherwise it is unchanged. Set Gamma_n(t)=Psi_n(K_n(t)). Extend this particular prescribed path to all real times by gamma(s)=Gamma_n(t)(min(max(s,0),T)). This extension agrees with Gamma on[0,T]; no topology on an entire function space is asserted. Define momentum reversal R(x,p)=(x,-p), target energy U_y(x,p)=V(x)+||x-y||^2/(2 eta)+||p||^2/2 and unnormalized Gibbs weight w_y(z)=exp(-U_y(z)). For any actual initial state z and strict event tuple t, let P_(y,r,z,n,T)(t)=product_(i=1)^n lambda_r(Phi_(a_i-a_(i-1))(z_(i-1))), I_(y,r,z,n,T)(t)=integral_0^T lambda_r(gamma(s)) ds and dReal_(y,r,z,n,T)(t)=exp(-I_(y,r,z,n,T)(t))*P_(y,r,z,n,T)(t). The incoming-rate product excludes the terminal arrival. The original ENNReal density rho_n(T,t) remains q_n times the terminal live survival factor defined above; it is not redefined as ofReal(dReal). For each n>=0 and T>=0 there exists a single reversal map rho:Prefix_n(T)->Prefix_n(T), with (rho(t))_i=T-t_(n+1-i) in one-based coordinates. For every y,r,z and strict tuple t, let A=Gamma_(y,r,z,n,T,t)(T). Then both identities hold: w_y(z) dReal_(y,r,z,n,T)(t) = w_y(R(A)) dReal_(y,r,R(A),n,T)(rho(t)), ofReal(w_y(z)) rho_(y,r,z,n,T)(t) = ofReal(w_y(R(A))) rho_(y,r,R(A),n,T)(rho(t)). The reversal map and all internal identities are produced, not assumed. The six original analytic assumptions are the only analytic premises. There is no positivity/nonzero-rate premise and no cancellation of event rates. The conclusions include n=0 with empty product1, T=0, zero-dimensional phase space, zero normals and zero rates; for T=0 and n>0 the strict event subtype is empty. This is unnormalized pointwise weighted density matching in Appendix A.1, not a weighted integral/sector identity, normalized invariant probability, Markov property, hypocoercivity, implementation error, actual expected cost, paper main result or PBPS/SPHMC composition.

Braces mark parameters Lean can infer; square brackets request structures such as a measurable space or probability measure. Named hypotheses are mathematical premises, not facts established by this declaration. Section parameters are described in the mathematical hypotheses above; the module link retains their exact source context.

theorem actual_prescribed_weighted_density_balance
    {E : Type*} [NormedAddCommGroup E] [InnerProductSpace ℝ E]
    [FiniteDimensional ℝ E] [MeasurableSpace E] [BorelSpace E]
    {V : E → ℝ} {α β : ℝ≥0} {η : ℝ}
    (hα : 0 < (α : ℝ)) (hαβ : α ≤ β) (hV : ContDiff ℝ 2 V)
    (hH : ∀ x v : E,
      (α : ℝ) * ‖v‖ ^ 2 ≤ (fderiv ℝ (fderiv ℝ V) x v) v ∧
      (fderiv ℝ (fderiv ℝ V) x v) v ≤ (β : ℝ) * ‖v‖ ^ 2)
    (hη : 0 < η) (hβη : (β : ℝ) * η ≤ 1) : actual_prescribed_weighted_density_statement hα hαβ hV hH hη hβη

Exact module and namespace context

Full Lean proposition (definition) · actual_prescribed_weighted_density_statement

This literal definition stores the complete proposition asserted above. It supplies no mathematical proof or additional hypothesis; the theorem proof must establish its value.

Braces mark parameters Lean can infer; square brackets request structures such as a measurable space or probability measure. Named hypotheses are mathematical premises, not facts established by this declaration. Section parameters are described in the mathematical hypotheses above; the module link retains their exact source context.

Download module (.lean)
private def actual_prescribed_weighted_density_statement
    {E : Type*} [NormedAddCommGroup E] [InnerProductSpace ℝ E]
    [FiniteDimensional ℝ E] [MeasurableSpace E] [BorelSpace E]
    {V : E → ℝ} {α β : ℝ≥0} {η : ℝ}
    (hα : 0 < (α : ℝ)) (hαβ : α ≤ β) (hV : ContDiff ℝ 2 V)
    (hH : ∀ x v : E,
      (α : ℝ) * ‖v‖ ^ 2 ≤ (fderiv ℝ (fderiv ℝ V) x v) v ∧
      (fderiv ℝ (fderiv ℝ V) x v) v ≤ (β : ℝ) * ‖v‖ ^ 2)
    (hη : 0 < η) (hβη : (β : ℝ) * η ≤ 1) : Prop := by
  classical
  exact
    let c : E → E → E := fun y xRef => y - η • gradient V xRef
    let Φ : E → E → ℝ → (E × E) → (E × E) := fun y xRef t z =>
      (c y xRef + Real.cos t • (z.1 - c y xRef) +
          (Real.sqrt η * Real.sin t) • z.2,
       (-Real.sin t / Real.sqrt η) • (z.1 - c y xRef) + Real.cos t • z.2)
    let S : E → (E × E) → (E × E) := fun xRef z =>
      (z.1, z.2 - (2 * inner ℝ z.2 (gradient V z.1 - gradient V xRef) /
        ‖gradient V z.1 - gradient V xRef‖ ^ 2) • (gradient V z.1 - gradient V xRef))
    let rate : E → (E × E) → ℝ := fun xRef z =>
      Real.sqrt η * max 0 (inner ℝ z.2 (gradient V z.1 - gradient V xRef))
    let Λ : E → E → (E × E) → ℝ≥0 → ℝ := fun y xRef z t =>
      ∫ s in (0 : ℝ)..(t : ℝ), rate xRef (Φ y xRef s z)
    let Rec := (ℝ≥0 × (E × E)) ⊕ Unit
    let Hist : ℕ → Type _ := fun n => Fin (n + 1) → Rec
    let Δ : (n : ℕ) → ℝ≥0 → Set (Fin n → ℝ) := fun n T =>
      {t | StrictMono (Fin.cons (0 : ℝ) t) ∧ ∀ i, t i < (T : ℝ)}
    let Prefix : (n : ℕ) → ℝ≥0 → Type := fun n T => {t : Fin n → ℝ // t ∈ Δ n T}
    let zAt : E → E → (E × E) → (n : ℕ) → (T : ℝ≥0) →
        Prefix n T → Fin (n + 1) → (E × E) := fun y xRef z₀ n T t =>
      Fin.induction z₀ (fun i z =>
        S xRef (Φ y xRef (t.val i - (Fin.cons (0 : ℝ) (t.val : Fin n → ℝ) : Fin (n + 1) → ℝ) i.castSucc) z))
    let K : E → E → (E × E) → (n : ℕ) → (T : ℝ≥0) →
        Prefix n T → Hist n := fun y xRef z₀ n T t j =>
      Sum.inl (Real.toNNReal ((Fin.cons (0 : ℝ) (t.val : Fin n → ℝ) : Fin (n + 1) → ℝ) j), zAt y xRef z₀ n T t j)
    let qReal : E → E → (E × E) → (n : ℕ) → (T : ℝ≥0) →
        Prefix n T → ℝ := fun y xRef z₀ n T t =>
      ∏ i : Fin n,
        rate xRef (Φ y xRef (t.val i - (Fin.cons (0 : ℝ) (t.val : Fin n → ℝ) : Fin (n + 1) → ℝ) i.castSucc)
          (zAt y xRef z₀ n T t i.castSucc)) *
        Real.exp (-Λ y xRef (zAt y xRef z₀ n T t i.castSucc)
          (Real.toNNReal (t.val i - (Fin.cons (0 : ℝ) (t.val : Fin n → ℝ) : Fin (n + 1) → ℝ) i.castSucc)))
    let q : E → E → (E × E) → (n : ℕ) → (T : ℝ≥0) →
        Prefix n T → ℝ≥0∞ := fun y xRef z₀ n T t =>
      ENNReal.ofReal (qReal y xRef z₀ n T t)
    let liveTime : Rec → ℝ≥0 := fun r => r.elim Prod.fst (fun _ => 0)
    let d : (n : ℕ) → ℝ≥0 → Hist n → ℝ≥0 :=
      fun n T h => T - liveTime (h ⟨n, Nat.lt_succ_self n⟩)
    let weight : E → E → Rec → ℝ≥0 → ℝ≥0∞ :=
      fun y xRef r u => r.elim
        (fun a => ENNReal.ofReal (Real.exp (-Λ y xRef a.2 u))) (fun _ => 0)
    let Path : ℝ≥0 → Type _ := fun T => {s : ℝ≥0 // s ≤ T} → E × E
    let Ψ : E → E → (E × E) → (n : ℕ) → (T : ℝ≥0) → Hist n → Path T :=
      fun y xRef z₀ n T h s =>
        Fin.foldl (n + 1) (fun z j => (h j).elim
          (fun a => if a.1 ≤ s.val then Φ y xRef ((s.val - a.1 : ℝ≥0) : ℝ) a.2 else z)
          (fun _ => z)) z₀
    let Γ : E → E → (E × E) → (n : ℕ) → (T : ℝ≥0) → Prefix n T → Path T :=
      fun y xRef z₀ n T t => Ψ y xRef z₀ n T (K y xRef z₀ n T t)
    let ρ : E → E → (E × E) → (n : ℕ) → (T : ℝ≥0) → Prefix n T → ℝ≥0∞ :=
      fun y xRef z₀ n T t => q y xRef z₀ n T t *
        weight y xRef ((K y xRef z₀ n T t) ⟨n, Nat.lt_succ_self n⟩)
          (d n T (K y xRef z₀ n T t))
    let R : E × E → E × E := fun z => (z.1, -z.2)
    let U : E → E × E → ℝ := fun y z =>
      V z.1 + ‖z.1 - y‖ ^ 2 / (2 * η) + ‖z.2‖ ^ 2 / 2
    let w : E → E × E → ℝ := fun y z => Real.exp (-U y z)
    let P : E → E → (E × E) → (n : ℕ) → (T : ℝ≥0) → Prefix n T → ℝ :=
      fun y r z n T t => ∏ i : Fin n,
        rate r (Φ y r (t.val i - (Fin.cons (0 : ℝ) (t.val : Fin n → ℝ) : Fin (n+1) → ℝ) i.castSucc)
          (zAt y r z n T t i.castSucc))
    let I : E → E → (E × E) → (n : ℕ) → (T : ℝ≥0) → Prefix n T → ℝ :=
      fun y r z n T t => ∫ s in (0 : ℝ)..(T : ℝ),
        rate r (Γ y r z n T t ⟨min (Real.toNNReal s) T, min_le_right _ _⟩)
    let dReal : E → E → (E × E) → (n : ℕ) → (T : ℝ≥0) → Prefix n T → ℝ :=
      fun y r z n T t => Real.exp (-I y r z n T t) * P y r z n T t
    ∀ n : ℕ, ∀ T : ℝ≥0, ∃ rho : Prefix n T → Prefix n T,
      (∀ t : Prefix n T, ∀ i : Fin n, (rho t).val i = (T : ℝ) - t.val i.rev) ∧
      ∀ y r : E, ∀ z : E × E, ∀ t : Prefix n T,
        let A : E × E := Γ y r z n T t ⟨T, le_rfl⟩
        w y z * dReal y r z n T t = w y (R A) * dReal y r (R A) n T (rho t) ∧
        ENNReal.ofReal (w y z) * ρ y r z n T t =
          ENNReal.ofReal (w y (R A)) * ρ y r (R A) n T (rho t)

Exact module and namespace context

Proof roadmap

  1. 1Choose one reversed event tuple

    Using zero-based Fin n coordinates (i=0,...,n-1), the path and state reversal producers each return a map with the same exact coordinate formula. Function and subtype extensionality identify the two maps; the resulting witness serves every parameter and initial state.

  2. 2Match the incoming event-rate product

    The finite state reversal result pairs incoming rates after reversing the event order. Simplifying the prepended/appended grid indices gives equality of the literal P products. Zero factors and the empty product remain valid.

  3. 3Identify the reversed survival integral

    Apply the actual off-event path reversal and its proved scalar integral substitution. Internal jumps form a finite null set; no pointwise equality at the jumps is needed in this integral identity.

  4. 4Combine target energy and survival

    The Gibbs survival producer gives w(z) exp(-Iplus)=w(A) exp(-Iminus). Momentum reversal leaves position and squared momentum norm unchanged, so U(RA)=U(A). Substitute the reversed integral from the preceding step.

  5. 5Attach the common rate product

    Multiply the survival identity by the incoming-rate product and reassociate multiplication. This proves the real weighted density identity without division or a positive-rate assumption.

  6. 6Recover the actual algorithmic density

    Apply the actual prescribed-density theorem at both inputs to rewrite q times terminal survival as ofReal(dReal). Apply ofReal to the real identity and distribute it over the products using nonnegativity of exp(-U). The final identity is about the original actual ENNReal density.

Full assumptions and notation

All objects and hypotheses

  • Finite-dimensional real inner-product Borel E, including rank zero.
  • alpha,beta≥0; alpha>0; alpha≤beta; V is C2; alpha||v||^2≤D2V(x)[v,v]≤beta||v||^2 for every x,v; eta>0; beta*eta≤1.
  • Every finite n≥0,T≥0; the same produced event-time reversal works for all y,r,z,t. Zero rates and normals are allowed.
  • Exact actual rho=q*terminalweight, actual finite flow/reflection recursion and clamped path; no supplied reversal, product, energy, density or integrability assumption.

Mathematical proof

1. Choose one reversed event tuple

Using zero-based Fin n coordinates (i=0,...,n-1), the path and state reversal producers each return a map with the same exact coordinate formula. Function and subtype extensionality identify the two maps; the resulting witness serves every parameter and initial state.

\[\rho(t)_i=T-t_{n-1-i},\quad\rho_{99}=\rho_{100}.\]
Corresponding Lean step
theorem actual_prescribed_weighted_density_balance
    {E : Type*} [NormedAddCommGroup E] [InnerProductSpace ℝ E]
    [FiniteDimensional ℝ E] [MeasurableSpace E] [BorelSpace E]
    {V : E → ℝ} {α β : ℝ≥0} {η : ℝ}
    (hα : 0 < (α : ℝ)) (hαβ : α ≤ β) (hV : ContDiff ℝ 2 V)
    (hH : ∀ x v : E,
      (α : ℝ) * ‖v‖ ^ 2 ≤ (fderiv ℝ (fderiv ℝ V) x v) v ∧
      (fderiv ℝ (fderiv ℝ V) x v) v ≤ (β : ℝ) * ‖v‖ ^ 2)
    (hη : 0 < η) (hβη : (β : ℝ) * η ≤ 1) : actual_prescribed_weighted_density_statement hα hαβ hV hH hη hβη := by
  classical
  unfold actual_prescribed_weighted_density_statement
  intro c Φ S rate Λ Rec Hist Δ Prefix zAt K qReal q liveTime d weight Path Ψ Γ ρ R U w P I dReal n T
  obtain ⟨rho, hrho, h100⟩ :=
    AutoSamplingTheory.ExampleCases.ProximalBPS.ActualPrescribedPathReversal.actual_prescribed_path_reversal_and_survival
      hα hαβ hV hH hη hβη n T
  obtain ⟨rho99, hrho99, h99⟩ :=
    AutoSamplingTheory.ExampleCases.ProximalBPS.ActualPrescribedStateReversal.actual_prescribed_state_reversal_and_rate_product
      hα hαβ hV hH hη hβη n T
  -- The public coordinate equations identify the independently produced witnesses.
  have halign : rho99 = rho := by
    funext t
    apply Subtype.ext
    funext i
    exact (hrho99 t i).trans (hrho t i).symm
  subst rho99
  refine ⟨rho, hrho, ?_⟩
  intro y r z t A

Exact compiled region. Each producer premise is discharged by the original assumptions; no proof ingredient is a new public binder.

2. Match the incoming event-rate product

The finite state reversal result pairs incoming rates after reversing the event order. Simplifying the prepended/appended grid indices gives equality of the literal P products. Zero factors and the empty product remain valid.

\[P_{R A,\rho(t)}=P_{z,t}.\]
Corresponding Lean step
  -- Incoming rates, including zero factors, pair under finite state reversal.
  have hprod : P y r (R A) n T (rho t) = P y r z n T t := by
    simpa only [P, Fin.succ_castSucc, Fin.snoc_castSucc, Fin.cons_succ] using
      (h99 y r z t).2

Exact compiled region. Each producer premise is discharged by the original assumptions; no proof ingredient is a new public binder.

3. Identify the reversed survival integral

Apply the actual off-event path reversal and its proved scalar integral substitution. Internal jumps form a finite null set; no pointwise equality at the jumps is needed in this integral identity.

\[I_{R A,\rho(t)}=\int_0^T\lambda_r(R\gamma(s))\,ds.\]
Corresponding Lean step
  -- The reversed survival integral equals the momentum-flipped forward integral.
  have hint : I y r (R A) n T (rho t) =
      ∫ s in (0 : ℝ)..(T : ℝ), rate r (R (Γ y r z n T t
        ⟨min (Real.toNNReal s) T, min_le_right _ _⟩)) :=
    (h100 y r z t).2.2.2

Exact compiled region. Each producer premise is discharged by the original assumptions; no proof ingredient is a new public binder.

4. Combine target energy and survival

The Gibbs survival producer gives w(z) exp(-Iplus)=w(A) exp(-Iminus). Momentum reversal leaves position and squared momentum norm unchanged, so U(RA)=U(A). Substitute the reversed integral from the preceding step.

\[w(z)e^{-I_{z,t}}=w(RA)e^{-I_{RA,\rho(t)}}.\]
Corresponding Lean step
  -- Gibbs survival balance and invariance of target energy under momentum flip.
  have hbal :=
    (AutoSamplingTheory.ExampleCases.ProximalBPS.ActualPrescribedGibbsSurvival.actual_prescribed_gibbs_survival_balance
      hα hαβ hV hH hη hβη y r z n T t).2.2.2
  have hw : w y (R A) = w y A := by simp only [w, U, R, norm_neg]
  have hs : w y z * Real.exp (-I y r z n T t) =
      w y (R A) * Real.exp (-I y r (R A) n T (rho t)) := by
    rw [hw, hint]
    exact hbal

Exact compiled region. Each producer premise is discharged by the original assumptions; no proof ingredient is a new public binder.

5. Attach the common rate product

Multiply the survival identity by the incoming-rate product and reassociate multiplication. This proves the real weighted density identity without division or a positive-rate assumption.

\[w(z)d^{\mathrm{real}}_{z,t}=w(RA)d^{\mathrm{real}}_{RA,\rho(t)}.\]
Corresponding Lean step
  -- Multiply by the common event product; no rate cancellation is used.
  have hreal : w y z * dReal y r z n T t =
      w y (R A) * dReal y r (R A) n T (rho t) := by
    dsimp only [dReal]
    rw [hprod, ← mul_assoc, ← mul_assoc, hs]
  refine ⟨hreal, ?_⟩

Exact compiled region. Each producer premise is discharged by the original assumptions; no proof ingredient is a new public binder.

6. Recover the actual algorithmic density

Apply the actual prescribed-density theorem at both inputs to rewrite q times terminal survival as ofReal(dReal). Apply ofReal to the real identity and distribute it over the products using nonnegativity of exp(-U). The final identity is about the original actual ENNReal density.

\[\operatorname{ofReal}(w(z))\rho_{z,t}=\operatorname{ofReal}(w(RA))\rho_{RA,\rho(t)}.\]
Corresponding Lean step
  -- Reconnect both real densities to the original algorithmic ENNReal density.
  have hd (a : E × E) (u : Prefix n T) :
      ρ y r a n T u = ENNReal.ofReal (dReal y r a n T u) := by
    simpa only [dReal, I, P, Fin.snoc_castSucc] using
      (AutoSamplingTheory.ExampleCases.ProximalBPS.ActualPrescribedPathDensity.actual_prescribed_path_density_rewrite
      hα hαβ hV hH hη hβη y r a n T u).2.2.2
  rw [hd z t, hd (R A) (rho t), ← ENNReal.ofReal_mul (Real.exp_nonneg _),
    ← ENNReal.ofReal_mul (Real.exp_nonneg _), hreal]

Exact compiled region. Each producer premise is discharged by the original assumptions; no proof ingredient is a new public binder.

Lean verification

The exact proof begins folded immediately after the mathematical argument; the statement is folded beside the claim above.

Lean proof · actual_prescribed_weighted_density_balance

Choose one reversed event tuple; Match the incoming event-rate product; Identify the reversed survival integral; Combine target energy and survival; Attach the common rate product; Recover the actual algorithmic density

Braces mark parameters Lean can infer; square brackets request structures such as a measurable space or probability measure. Named hypotheses are mathematical premises, not facts established by this declaration. Section parameters are described in the mathematical hypotheses above; the module link retains their exact source context.

Download module (.lean)
theorem actual_prescribed_weighted_density_balance
    {E : Type*} [NormedAddCommGroup E] [InnerProductSpace ℝ E]
    [FiniteDimensional ℝ E] [MeasurableSpace E] [BorelSpace E]
    {V : E → ℝ} {α β : ℝ≥0} {η : ℝ}
    (hα : 0 < (α : ℝ)) (hαβ : α ≤ β) (hV : ContDiff ℝ 2 V)
    (hH : ∀ x v : E,
      (α : ℝ) * ‖v‖ ^ 2 ≤ (fderiv ℝ (fderiv ℝ V) x v) v ∧
      (fderiv ℝ (fderiv ℝ V) x v) v ≤ (β : ℝ) * ‖v‖ ^ 2)
    (hη : 0 < η) (hβη : (β : ℝ) * η ≤ 1) : actual_prescribed_weighted_density_statement hα hαβ hV hH hη hβη := by
  classical
  unfold actual_prescribed_weighted_density_statement
  intro c Φ S rate Λ Rec Hist Δ Prefix zAt K qReal q liveTime d weight Path Ψ Γ ρ R U w P I dReal n T
  obtain ⟨rho, hrho, h100⟩ :=
    AutoSamplingTheory.ExampleCases.ProximalBPS.ActualPrescribedPathReversal.actual_prescribed_path_reversal_and_survival
      hα hαβ hV hH hη hβη n T
  obtain ⟨rho99, hrho99, h99⟩ :=
    AutoSamplingTheory.ExampleCases.ProximalBPS.ActualPrescribedStateReversal.actual_prescribed_state_reversal_and_rate_product
      hα hαβ hV hH hη hβη n T
  -- The public coordinate equations identify the independently produced witnesses.
  have halign : rho99 = rho := by
    funext t
    apply Subtype.ext
    funext i
    exact (hrho99 t i).trans (hrho t i).symm
  subst rho99
  refine ⟨rho, hrho, ?_⟩
  intro y r z t A
  -- Incoming rates, including zero factors, pair under finite state reversal.
  have hprod : P y r (R A) n T (rho t) = P y r z n T t := by
    simpa only [P, Fin.succ_castSucc, Fin.snoc_castSucc, Fin.cons_succ] using
      (h99 y r z t).2
  -- The reversed survival integral equals the momentum-flipped forward integral.
  have hint : I y r (R A) n T (rho t) =
      ∫ s in (0 : ℝ)..(T : ℝ), rate r (R (Γ y r z n T t
        ⟨min (Real.toNNReal s) T, min_le_right _ _⟩)) :=
    (h100 y r z t).2.2.2
  -- Gibbs survival balance and invariance of target energy under momentum flip.
  have hbal :=
    (AutoSamplingTheory.ExampleCases.ProximalBPS.ActualPrescribedGibbsSurvival.actual_prescribed_gibbs_survival_balance
      hα hαβ hV hH hη hβη y r z n T t).2.2.2
  have hw : w y (R A) = w y A := by simp only [w, U, R, norm_neg]
  have hs : w y z * Real.exp (-I y r z n T t) =
      w y (R A) * Real.exp (-I y r (R A) n T (rho t)) := by
    rw [hw, hint]
    exact hbal
  -- Multiply by the common event product; no rate cancellation is used.
  have hreal : w y z * dReal y r z n T t =
      w y (R A) * dReal y r (R A) n T (rho t) := by
    dsimp only [dReal]
    rw [hprod, ← mul_assoc, ← mul_assoc, hs]
  refine ⟨hreal, ?_⟩
  -- Reconnect both real densities to the original algorithmic ENNReal density.
  have hd (a : E × E) (u : Prefix n T) :
      ρ y r a n T u = ENNReal.ofReal (dReal y r a n T u) := by
    simpa only [dReal, I, P, Fin.snoc_castSucc] using
      (AutoSamplingTheory.ExampleCases.ProximalBPS.ActualPrescribedPathDensity.actual_prescribed_path_density_rewrite
      hα hαβ hV hH hη hβη y r a n T u).2.2.2
  rw [hd z t, hd (R A) (rho t), ← ENNReal.ofReal_mul (Real.exp_nonneg _),
    ← ENNReal.ofReal_mul (Real.exp_nonneg _), hreal]

#print axioms actual_prescribed_weighted_density_balance
end
end AutoSamplingTheory.ExampleCases.ProximalBPS.ActualPrescribedWeightedDensity

Exact module and namespace context

Scope and omitted-condition boundaries
  • Pointwise unnormalized weighted-density matching only; no weighted integral/sector, invariant probability, restart/Markovness, hypocoercivity, implementation error, expected cost, main theorem or composition.
Sources, dependencies, and reuse

Source and reuse

Samplinglib parents called

Mathlib API called (external library)

  • Subtype.ext
  • Fin.snoc_castSucc
  • Fin.cons_succ
  • norm_neg
  • mul_assoc
  • Real.exp_nonneg
  • ENNReal.ofReal_mul

Mathematical sources

Samplinglib prose is not a quotation or a source-equivalence certificate. Definitions and aliases are explained as constructions, not counted as new mathematical proofs.