Theorem 2, C3 — Eventual validity and Gaussianity of the DKF update
ProvedDKFAsymptotics.C3Let the state dimension be positive and finite. Let be real and let be symmetric positive-definite matrices with . Set and .
Let be probability densities and let be Gaussian probability densities with positive-definite covariances, for each positive integer . Assume the normalized density is well defined for every . Let be an arbitrary probability measure and . Assume
All densities and distances use Lebesgue measure. Weak convergence means convergence against every bounded continuous real test function. The approximate update is the specific expression ; its validity is not assumed at every index.
Then, for all sufficiently large , the approximate update has a measurable weight and a finite, strictly positive normalizer, and is a Gaussian probability density with positive-definite covariance.
import Definitions.Def_DKFAsymptotics_Assumptions noncomputable section open MeasureTheory Filter Matrix open scoped Topology ENNReal
namespace DKFAsymptotics
theorem C3 {d : ℕ} [NeZero d] (M : Model d) (D : Problem M)
(h : Assumptions M D) :
∀ᶠ n in atTop, UpdateDefined M (D.s' n) (D.u' n) ∧
IsGaussianPDF (D.p' M n) := by sorry
end DKFAsymptoticsRead-back
What the Lean code literally says, in plain math · GPT-6 (Codex independent sub-agent)
Read-back of C3
For every integer , identify the state space with (coordinates indexed by ), equipped with its product Lebesgue measure . Let be real matrices such that and are symmetric positive definite and ; here symmetric positive definite means and for every nonzero , and no separate positivity, symmetry, or invertibility assumption is placed on . For a mean and a real matrix , put . A real function on is called a probability density here precisely when it is nonnegative -almost everywhere, is Lebesgue integrable, and satisfies ; it is called a Gaussian probability density precisely when it is a probability density and there exist and a symmetric positive definite such that -almost everywhere. For any real function , define its extended nonnegative norm by , and define the measure on measurable sets by , with this last integral understood as the extended nonnegative Lebesgue integral. For real functions , put , , and , where maps a finite extended nonnegative real to the corresponding real and maps to ; real division by zero is defined to give zero, and a real Bochner integral is defined to give zero if its integrand is not integrable. Say that the update for is defined precisely when is almost everywhere equal to a strongly measurable real function and . Let be four sequences of real functions on , let , and let be any probability measure on . Assume that for every positive integer , both and are probability densities as just defined, both and are Gaussian probability densities as just defined, and the update for is defined. Write and . Assume all five of the following convergence conditions as tends to infinity through the positive integers: for every bounded continuous real function on , ; in the topology of ; for every such , , where is the unit point mass at ; in the same extended nonnegative real topology; and for every such , . Then there exists a positive integer such that, for every positive integer , the update for is defined, and is a Gaussian probability density: explicitly, is almost everywhere equal to a strongly measurable function, , is nonnegative almost everywhere and integrable with , and there exist and a symmetric positive definite real matrix with for -almost every . The Gaussian representations are asserted only almost everywhere and are not asserted to be unique; their parameters may depend on . The index and dimension are excluded. The definition of applies to every positive index even when its normalizing norm is zero or infinite, in which cases it gives the zero function; the conclusion rules out those normalizing cases only for and asserts no such conclusion at earlier indices.
Confirmed by the mission captain (proposal self-audit).