Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 2, C4 — Concentration of the DKF update

Proved
DKFAsymptotics.C4

by burkh4rt · Sep 15, 2026 · Mathlib 0df444a (Lean v4.33.1)

asymptoticskalman-filterprobabilitystatistics

Let the state dimension be positive and finite. Let AAA be real and let Γ,S\Gamma,SΓ,S be symmetric positive-definite matrices with S=ASA⊤+ΓS=ASA^\top+\GammaS=ASA⊤+Γ. Set p(z)=ηd(z;0,S)p(z)=\eta_d(z;0,S)p(z)=ηd​(z;0,S) and (τs)(z)=∫ηd(z;Ay,Γ)s(y) dy(\tau s)(z)=\int\eta_d(z;Ay,\Gamma)s(y)\,dy(τs)(z)=∫ηd​(z;Ay,Γ)s(y)dy.

Let sn,uns_n,u_nsn​,un​ be probability densities and let sn′,un′s'_n,u'_nsn′​,un′​ be Gaussian probability densities with positive-definite covariances, for each positive integer nnn. Assume the normalized density pn=(unτsn/p)/∥unτsn/p∥1p_n=(u_n\tau s_n/p)/\|u_n\tau s_n/p\|_1pn​=(un​τsn​/p)/∥un​τsn​/p∥1​ is well defined for every nnn. Let PPP be an arbitrary probability measure and b∈Rdb\in\mathbb R^db∈Rd. Assume

sn⇒P,∥sn−sn′∥1→0,un⇒δb,∥un−un′∥1→0,pn⇒δb.s_n\Rightarrow P,\quad \|s_n-s'_n\|_1\to0,\quad u_n\Rightarrow\delta_b,\quad \|u_n-u'_n\|_1\to0,\quad p_n\Rightarrow\delta_b.sn​⇒P,∥sn​−sn′​∥1​→0,un​⇒δb​,∥un​−un′​∥1​→0,pn​⇒δb​.

All densities and L1L^1L1 distances use Lebesgue measure. Weak convergence means convergence against every bounded continuous real test function. The approximate update is the specific expression pn′=(un′τsn′/p)/∥un′τsn′/p∥1p'_n=(u'_n\tau s'_n/p)/\|u'_n\tau s'_n/p\|_1pn′​=(un′​τsn′​/p)/∥un′​τsn′​/p∥1​; its validity is not assumed at every index.

Then pn′⇒δbp'_n\Rightarrow\delta_bpn′​⇒δb​. The conclusion concerns the same normalized update as C3, including its eventual validity established in that milestone.

Preamble
import Definitions.Def_DKFAsymptotics_Assumptions

noncomputable section
open MeasureTheory Filter Matrix
open scoped Topology ENNReal
Formal statement
namespace DKFAsymptotics

theorem C4 {d : ℕ} [NeZero d] (M : Model d) (D : Problem M)
    (h : Assumptions M D) :
    Weakly (D.p' M) (Measure.dirac D.b) := by sorry

end DKFAsymptotics
Source
M. C. Burkhart, D. M. Brandman, B. Franco, L. R. Hochberg and M. T. Harrison, The Discriminative Kalman Filter for Bayesian Filtering with Nonlinear and Nongaussian Observation Models, Neural Computation 32(5), 969–1017 (2020), doi:10.1162/neco_a_01275; https://burkh4rt.github.io/pubs/Burkhart-et-al-2020-Neural-Comput.pdf; Appendix B, Theorem 2, conclusion C4, p. 1003; proof pp. 1005–1007 and Lemma 1, pp. 1007–1010.
Read-back

What the Lean code literally says, in plain math · GPT-6 (Codex independent sub-agent)

Read-back of DKFAsymptotics.C4

For every natural number d≥1d\geq 1d≥1, identify the state space with Rd\mathbb R^dRd, with its usual product topology, Borel measurable sets, and product Lebesgue measure λ\lambdaλ. Let A,Γ,SA,\Gamma,SA,Γ,S be real d×dd\times dd×d matrices such that Γ\GammaΓ and SSS are symmetric positive definite (that is, xTCx>0x^\mathsf TCx>0xTCx>0 for every nonzero x∈Rdx\in\mathbb R^dx∈Rd, for C=Γ,SC=\Gamma,SC=Γ,S) and S=ASAT+ΓS=ASA^\mathsf T+\GammaS=ASAT+Γ; there is no separate symmetry or positivity requirement on AAA. For a vector mmm and a symmetric positive definite matrix CCC, put ϕm,C(z)=((2π)ddet⁡C)−1exp⁡(−(z−m)TC−1(z−m)/2)\phi_{m,C}(z)=\bigl(\sqrt{(2\pi)^d\det C}\bigr)^{-1}\exp\bigl(- (z-m)^\mathsf TC^{-1}(z-m)/2\bigr)ϕm,C​(z)=((2π)ddetC​)−1exp(−(z−m)TC−1(z−m)/2). Let sn,un,sn′,un′:Rd→Rs_n,u_n,s'_n,u'_n:\mathbb R^d\to\mathbb Rsn​,un​,sn′​,un′​:Rd→R, for every positive integer nnn, be four sequences of functions, let b∈Rdb\in\mathbb R^db∈Rd, and let PPP be any Borel probability measure on Rd\mathbb R^dRd. Every one of the four functions at each index is assumed to be nonnegative λ\lambdaλ-almost everywhere, Lebesgue integrable, and to have integral 111. In addition, for every nnn there exist vectors mns,mnu∈Rdm_n^s,m_n^u\in\mathbb R^dmns​,mnu​∈Rd and symmetric positive definite matrices Cns,CnuC_n^s,C_n^uCns​,Cnu​ such that sn′=ϕmns,Cnss'_n=\phi_{m_n^s,C_n^s}sn′​=ϕmns​,Cns​​ and un′=ϕmnu,Cnuu'_n=\phi_{m_n^u,C_n^u}un′​=ϕmnu​,Cnu​​ λ\lambdaλ-almost everywhere; these identities and all nonnegativity assumptions allow exceptions on null sets. For any real-valued function fff, define its extended nonnegative integral N(f)=∫−∣f(z)∣ dλ(z)∈[0,∞]N(f)=\int^- |f(z)|\,d\lambda(z)\in[0,\infty]N(f)=∫−∣f(z)∣dλ(z)∈[0,∞], where the absolute value is viewed as an extended nonnegative number. Write rRr_{\mathbb R}rR​ for the ordinary real value of a finite r∈[0,∞]r\in[0,\infty]r∈[0,∞], with (+∞)R=0(+\infty)_{\mathbb R}=0(+∞)R​=0. Define qs(z)=∫ϕAy,Γ(z)s(y) dλ(y)q_s(z)=\int\phi_{Ay,\Gamma}(z)s(y)\,d\lambda(y)qs​(z)=∫ϕAy,Γ​(z)s(y)dλ(y), ws,u(z)=u(z)qs(z)/ϕ0,S(z)w_{s,u}(z)=u(z)q_s(z)/\phi_{0,S}(z)ws,u​(z)=u(z)qs​(z)/ϕ0,S​(z), and ps,u(z)=ws,u(z)/N(ws,u)Rp_{s,u}(z)=w_{s,u}(z)/N(w_{s,u})_{\mathbb R}ps,u​(z)=ws,u​(z)/N(ws,u​)R​, and abbreviate pn=psn,unp_n=p_{s_n,u_n}pn​=psn​,un​​ and pn′=psn′,un′p'_n=p_{s'_n,u'_n}pn′​=psn′​,un′​​. For every positive integer nnn, assume that wsn,unw_{s_n,u_n}wsn​,un​​ is almost everywhere equal to a strongly measurable function and that 0<N(wsn,un)<+∞0<N(w_{s_n,u_n})<+\infty0<N(wsn​,un​​)<+∞. For any real-valued fff, let μf\mu_fμf​ be the measure whose value on a Borel set EEE is μf(E)=∫E−max⁡{f(z),0} dλ(z)\mu_f(E)=\int^-_E\max\{f(z),0\}\,d\lambda(z)μf​(E)=∫E−​max{f(z),0}dλ(z), again viewing the nonnegative integrand as extended nonnegative. Suppose the following five conditions hold as n→∞n\to\inftyn→∞ through the positive integers: for every bounded continuous real-valued function ggg on Rd\mathbb R^dRd, ∫g dμsn→∫g dP\int g\,d\mu_{s_n}\to\int g\,dP∫gdμsn​​→∫gdP; N(sn−sn′)→0N(s_n-s'_n)\to0N(sn​−sn′​)→0 in the usual topology of [0,∞][0,\infty][0,∞]; for every such ggg, ∫g dμun→g(b)\int g\,d\mu_{u_n}\to g(b)∫gdμun​​→g(b); N(un−un′)→0N(u_n-u'_n)\to0N(un​−un′​)→0 in that same extended nonnegative topology; and for every such ggg, ∫g dμpn→g(b)\int g\,d\mu_{p_n}\to g(b)∫gdμpn​​→g(b). Then, for every bounded continuous real-valued ggg on Rd\mathbb R^dRd, ∫g dμpn′→g(b)\int g\,d\mu_{p'_n}\to g(b)∫gdμpn′​​→g(b) as n→∞n\to\inftyn→∞ through the positive integers. Here the measure concentrated at bbb has mass one, so its integral of ggg is g(b)g(b)g(b). The normalization formula defining ps,up_{s,u}ps,u​ is total: division by zero is assigned the value zero, so if N(ws,u)N(w_{s,u})N(ws,u​) is zero or infinite then ps,up_{s,u}ps,u​ is identically zero. The hypotheses explicitly require positive finite normalization for the pairs (sn,un)(s_n,u_n)(sn​,un​) only; they contain no corresponding requirement for the pairs (sn′,un′)(s'_n,u'_n)(sn′​,un′​). All real-valued integrals appearing in these definitions and convergence statements use the Bochner integral convention that a nonintegrable function has integral zero. Dimension zero and index zero are excluded, and no density or nondegeneracy condition is imposed on the probability measure PPP.

Human review
  • Endorsed by Shuze Chen · Sep 16, 2026

  • Endorsed by burkh4rt · Sep 16, 2026

    Confirmed by the mission captain (proposal self-audit).

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me