How statement and proof provenance work
The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.
- Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
- AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
- AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.
These labels describe origin, not correctness: citations and verification chips remain separate evidence.
ULA invariant laws have uniform exponential moments on the needed step ranges
Statement
Assume AC and the hypotheses of ULA's Gaussian kernel satisfies an exponential drift bound. Let be the ULA kernel, , , and . For every there is a unique invariant probability , and . There are finite constants , independent of , such that
Facts & Assumptions
Given: AC, the stated potential and radial constants, and .
AC is inherited through the Gaussian and weighted Hahn/Jordan interfaces. The Axiom of Choice
The ULA kernel has a positive continuous Gaussian density, with proposal . Unadjusted Langevin algorithm (ULA) The ULA Gaussian proposal is a probability kernel with a positive density
The fixed-step Gaussian kernel satisfies for a finite constant and a compact ball. ULA's Gaussian kernel satisfies an exponential drift bound
Radial drift gives for some ; hence V-sublevel sets are contained in compact balls. Radial gradient drift forces potential coercivity
Global geometric drift and minorization on a sufficiently large weight-sublevel set give invariant existence, a finite weight moment, and uniqueness among all invariant probabilities. Global drift and one-step minorization give weighted contraction and a unique invariant law
An exponential with positive exponent dominates every fixed nonnegative integer power at infinity. The exponential dominates every fixed nonnegative integer power at
Fermat gives a zero gradient at the minimizer. The chain rule and the fundamental theorem integrate the derivative along a segment, and Euclidean Cauchy–Schwarz bounds its inner products. Fermat's theorem: an interior differentiable local extremum has zero gradient The chain rule for total derivatives: Newton–Leibniz needs only continuity on , differentiability on , and a Riemann-integrable extension of the interior derivative Cauchy-Schwarz with its equality case, the triangle inequality for , the parallelogram law and polarisation
Standard Gaussian density, the scalar Gaussian integral, change of variables, Lebesgue product identification and Tonelli evaluate nonnegative finite-dimensional Gaussian integrals. A positive-definite multivariate normal law has its Gaussian density The Gaussian integral A C^1 diffeomorphism satisfies the change-of-variables formula for nonnegative Lebesgue measurable functions On Borel subsets of R^{m+n}, the product lambda_m times lambda_n agrees with lambda_{m+n} Tonelli's theorem for nonnegative measurable functions on a sigma-finite product
Proof
Given: The hypotheses in the Statement.
Fix . Dropping the indicator in [F2] gives a global drift with finite . Each sufficiently large V-sublevel is contained in a compact ball by [F3]. On the product of that starting ball and any fixed positive-radius target ball, the continuous positive density [F1] has a positive minimum: explicitly its Gaussian exponent is bounded because both balls and the continuous gradient are bounded. Restriction to the target ball and normalization of its positive finite volume give a common minorization. These are precisely the drift and minorization hypotheses for the weight V.
By [F6], , and therefore Lipschitz continuity gives . If is positive and at least the radial radius, Cauchy–Schwarz and the radial hypothesis give , so . For , integrate the derivative along and subtract to obtain
Completing the square, [F7] gives for and scalar v the identity Finite multiplication and Tonelli therefore give for .
Apply [F4] to the fixed-step kernel and weight V verified in step 1.1. It supplies a unique invariant probability with finite V moment.
Put and for with . Apply step 1.2 to , multiply by s, and exponentiate. Step 1.3, with quadratic coefficient and linear coefficient , gives The coefficient follows by combining with over the common denominator. All Gaussian integrals are finite under .
For , step 2.2 reduces to If , then , since for . Choose a radius R at least the radial radius and large enough that . Outside this ball step 1.2 makes the exponent at most . Inside it let . Since for , the preceding bound gives uniformly for small steps. Integrate against the finite-V invariant law from step 2.1 to obtain , using . This proves the first uniform bound.
For and , the logarithm of the prefactor in step 2.2 is at most , by the same integral estimate with . Moreover , because the numerator is at least and the denominator is at most one. Choose R' at least the radial radius such that . Outside this ball ; inside use and the exponential-difference estimate from step 3.1 with and upper horizon . Thus for the whole step range. Since , its invariant moment is finite by step 2.1. Integrating and using gives , uniformly for every allowed step.
Put . For , [F3] gives because . By [F5], is bounded on ; on use . Thus for some finite A. The Lipschitz bound in step 1.2 gives . Taking , and proves all asserted estimates. All constants depend only on the fixed potential and its displayed parameters.
Depends on
- The Axiom of Choice
- Unadjusted Langevin algorithm (ULA)
- The ULA Gaussian proposal is a probability kernel with a positive density
- ULA's Gaussian kernel satisfies an exponential drift bound
- Radial gradient drift forces potential coercivity
- Global drift and one-step minorization give weighted contraction and a unique invariant law
- The exponential dominates every fixed nonnegative integer power at $+\infty$
- Fermat's theorem: an interior differentiable local extremum has zero gradient
- The chain rule for total derivatives: $D(g\circ f)(a)=Dg(f(a))\circ Df(a)$
- Newton–Leibniz needs only continuity on $[a,b]$, differentiability on $(a,b)$, and a Riemann-integrable extension of the interior derivative
- Cauchy-Schwarz $\lvert\langle x,y\rangle\rvert \le \lVert x\rVert_2\lVert y\rVert_2$ with its equality case, the triangle inequality for $\lVert\cdot\rVert_2$, the parallelogram law and polarisation
- A positive-definite multivariate normal law has its Gaussian density
- The Gaussian integral $\int_{-\infty}^{\infty}e^{-x^2}\,dx=\sqrt{\pi}$
- A C^1 diffeomorphism satisfies the change-of-variables formula for nonnegative Lebesgue measurable functions
- On Borel subsets of R^{m+n}, the product lambda_m times lambda_n agrees with lambda_{m+n}
- Tonelli's theorem for nonnegative measurable functions on a sigma-finite product
Used by
Dependency tree · two levels
97 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.