Research map / Optimization / Nonconvex and smooth optimization
Wasserstein distance estimates for the distributions of numerical approximations to ergodic stochastic differential equations
Jesus Maria Sanz-Serna, Konstantinos C. Zygalakis
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use the UBU integrator instead of EE | underdamped Langevin discretization, c=1/L · kappa=1e9, h=2 | contractivity (not-contractive entries marked ***): *** -> 5.000(-10) | Improved |
| use the UBU integrator instead of EE | underdamped Langevin discretization, c=2/(L+m) · kappa=1e9, h=1 | contractivity (not-contractive entries marked ***): *** -> 1.000(-9) | Improved |
| use the UBU integrator instead of EE | underdamped Langevin discretization, c=3/(L+m) · kappa=1e9, h=1/2 | contractivity (not-contractive entries marked ***): *** -> 1.500(-9) | Improved |
| increase c from 1/L to 3/(L+m) | EE integrator · kappa=1e9, h=1 | contractivity (not-contractive entries marked ***): 5.000(-10) -> *** | Hurt |
| use smaller step size h=1/4 instead of h=1 | EE integrator, c=3/(L+m) · kappa=1e9 | contractivity (not-contractive entries marked ***): *** -> 1.500(-9) | Improved |
| increase c from 1/L to 2/(L+m) | UBU integrator, h=1/4 · kappa=1e9 | contractivity (not-contractive entries marked ***): 5.000(-10) -> 1.000(-9) | Improved |
| increase condition number kappa from 1e3 to 1e9 | contraction rate of both integrators · tables for kappa=1e3, 1e6, 1e9 | contractivity (not-contractive entries marked ***): multiplied by 10^6 or 10^3 | Hurt |
Closest papers by meaning
- Correction to "Wasserstein distance estimates for the distributions of numerical approximations to ergodic stochastic differential equations" (JMLR 2024 · similarity 0.93)
- Implicit Langevin Algorithms for Sampling From Log-concave Densities (JMLR 2021 · similarity 0.90)
- Wasserstein Convergence of Critically Damped Langevin Diffusions (NeurIPS 2025 · similarity 0.88)
- Continuous-time Riemannian SGD and SVRG Flows on Wasserstein Probabilistic Space (NeurIPS 2025 · similarity 0.87)
- Generalization of noisy SGD in unbounded non-convex settings (ICML 2025 · similarity 0.87)
- Assessing the quality of denoising diffusion models in Wasserstein distance: noisy score and optimal bounds (NeurIPS 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
