REVIEW 5 cited by
Convergence of kinetic Langevin samplers for non-convex potentials
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
abstract
We study three kinetic Langevin samplers including the Euler discretization, the BU and the UBU splitting scheme. We provide contraction results in $L^1$-Wasserstein distance for non-convex potentials. These results are based on a carefully tailored distance function and an appropriate coupling construction. Additionally, the error in the $L^1$-Wasserstein distance between the true target measure and the invariant measure of the discretization scheme is bounded. To get an $\varepsilon$-accuracy in $L^1$-Wasserstein distance, we show complexity guarantees of order $\mathcal{O}(\sqrt{d}/\varepsilon)$ for the Euler scheme and $\mathcal{O}(d^{1/4}/\sqrt{\varepsilon})$ for the UBU scheme under appropriate assumptions on the target measure. The results are applicable to interacting particle systems and provide bounds for sampling probability measures of mean-field type.
Forward citations
Cited by 5 Pith papers
-
Delocalization of bias in unadjusted Hamiltonian Monte Carlo and underdamped Langevin
Unadjusted HMC and BAOAB Langevin exhibit delocalization of bias: W2,ℓ∞ bias scales as O(h√log d) under weak/sparse interactions, so O(√K) integration steps control K-marginal bias.
-
Analysis of kinetic Langevin Monte Carlo under the stochastic exponential Euler discretization from underdamped all the way to overdamped
With the step size accelerated as h = h_LMC γ, the exponential-integrator kinetic Langevin Monte Carlo remains contractive in the overdamped limit, with contraction and bias matching Euler–Maruyama LMC.
-
Multimodal sampling via Schr\"odinger-F\"ollmer samplers with temperatures
Euler-discretized Schrödinger–Föllmer samplers with a temperature parameter provably converge at order O(h) in L2-Wasserstein distance, and high temperatures markedly improve multimodal sampling in experiments.
-
Discretization, Uniform-in-Time Estimations and Approximation of Invariant Measures for Nonlinear Stochastic Differential Equations with Non-Uniform Dissipativity
The numerical invariant measure of an explicit truncated Euler–Maruyama scheme converges to the true invariant measure at order h^{1/2} in the L1-Wasserstein distance, uniformly in time, for drifts that are dissipativ...
-
An explicit splitting SAV scheme for the kinetic Langevin dynamics
An explicit SSAV discretization of kinetic Langevin dynamics achieves order-one strong and weak convergence with error constants polynomial in time, even for superlinear-gradient potentials.
Discussion (0). Continue with ORCID to comment.