paper-with-me

Papers

Improved Bounds for Discretization of Langevin Diffusions: Near-Optimal Rates without Convexity

2019-07-25 · Wenlong Mou, Nicolas Flammarion, Martin J. Wainwright, Peter L. Bartlett

We present an improved analysis of the Euler-Maruyama discretization of the Langevin diffusion. Our analysis does not require global contractivity, and yields polynomial dependence on the time horizon. Compared to existing approaches, we make an additional smoothness assumption, and improve the existing rate from $O(\eta)$ to $O(\eta^2)$ in terms of the KL divergence. This result matches the correct order for numerical SDEs, without suffering from exponential time dependence. When applied to algorithms for sampling and learning, this result simultaneously improves all those methods based on Dalayan's approach.

📄 PDF Abstract BibTeX arXiv:1907.11331

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Global Non-convex Optimization with Discretized Diffusions

2018-10-29 · NeurIPS 2018 12 · Murat A. Erdogdu, Lester Mackey, Ohad Shamir

An Euler discretization of the Langevin diffusion is known to converge to the global minimizers of certain convex and non-convex optimization problems. We show that this property holds for any suitably smooth diffusion a…

Towards a Complete Analysis of Langevin Monte Carlo: Beyond Poincaré Inequality

2023-03-07 · Alireza Mousavi-Hosseini, Tyler Farghly, Ye He, Krishnakumar Balasubramanian 외

Langevin diffusions are rapidly convergent under appropriate functional inequality assumptions. Hence, it is natural to expect that with additional smoothness conditions to handle the discretization errors, their discret…

Stochastic Zeroth-order Discretizations of Langevin Diffusions for Bayesian Inference

2019-02-04 · Abhishek Roy, Lingqing Shen, Krishnakumar Balasubramanian, Saeed Ghadimi

Discretizations of Langevin diffusions provide a powerful method for sampling and Bayesian inference. However, such discretizations require evaluation of the gradient of the potential function. In several real-world scen…

Bayesian InferenceStochastic OptimizationVariable Selection

On sampling from a log-concave density using kinetic Langevin diffusions

2018-07-24 · Arnak S. Dalalyan, Lionel Riou-Durand

Langevin diffusion processes and their discretizations are often used for sampling from a target density. The most convenient framework for assessing the quality of such a sampling scheme corresponds to smooth and strong…

Wasserstein Control of Mirror Langevin Monte Carlo

2020-02-11 · Kelvin Shuangjian Zhang, Gabriel Peyré, Jalal Fadili, Marcelo Pereyra

Discretized Langevin diffusions are efficient Monte Carlo methods for sampling from high dimensional target densities that are log-Lipschitz-smooth and (strongly) log-concave. In particular, the Euclidean Langevin Monte …