PreprintarXiv licencearXiv
Accelerated High-Accuracy Sampling from a Warm Start via the Proximal Bouncy Particle Sampler
Abstract
We study the problem of sampling from μ(dx)∝ e−V(x) dx on ℝd, where V is α-strongly convex and β-smooth, and write κ:=β/α. We design and analyze the Proximal Bouncy Particle Sampler (Proximal BPS), a new sampler that combines ideas from the proximal sampler and the bouncy particle sampler. From a warm start initialization with $ O(1) $ Rényi divergence w.r.t. μ, Proximal BPS returns a sample whose law is ε-close to μ in total variation distance using Õ(√κ d1/4 polylog(1/ε)) gradient queries in expectation.
Related
- Reflected Proximal Dynamics for Accelerated High-Accuracy Log-Concave Sampling
- Local Geometric Mixing via Dobrushin Contraction with Applications to Diffusion Path Monte Carlo and the Proximal Sampler
- Entropy hypocoercivity and query complexity for the bouncy particle sampler
- Wasserstein accelerated proximal gradient algorithms
- Accelerated Proximal Gradient Method for Non-Strongly Convex Time-Varying Optimization
BibTeX
@misc{chen2026accelerated,
title = {{Accelerated High-Accuracy Sampling from a Warm Start via the Proximal Bouncy Particle Sampler}},
author = {Fan Chen and Sinho Chewi and Jianfeng Lu and Matthew S Zhang},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.06905},
eprint = {2609.06905},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.06905},
}
RIS
TY - UNPB
TI - Accelerated High-Accuracy Sampling from a Warm Start via the Proximal Bouncy Particle Sampler
AU - Fan Chen
AU - Sinho Chewi
AU - Jianfeng Lu
AU - Matthew S Zhang
PY - 2026
DA - 2026/09/07
PB - arXiv
DO - 10.48550/arxiv.2609.06905
UR - https://arxiv.org/abs/2609.06905
AB - We study the problem of sampling from μ(dx)∝ e^−V(x) dx on ℝ^d, where V is α-strongly convex and β-smooth, and write κ:=β/α. We design and analyze the Proximal Bouncy Particle Sampler (Proximal BPS), a new sampler that combines ideas from the proximal sampler and the bouncy particle sampler. From a warm start initialization with $ O(1) $ Rényi divergence w.r.t. μ, Proximal BPS returns a sample whose law is ε-close to μ in total variation distance using Õ(√κ d^1/4 polylog(1/ε)) gradient queries in expectation.
ER -