Tamed Langevin sampling under weaker conditions

• URL: http://arxiv.org/abs/2405.17693v1
• Date: Mon, 27 May 2024 23:00:40 GMT
• Title: Tamed Langevin sampling under weaker conditions
• Authors: Iosif Lytras, Panayotis Mertikopoulos,
• Abstract summary: We investigate the problem of sampling from distributions that are not log-concave and are only weakly dissipative. We introduce a taming scheme which is tailored to the growth and decay properties of the target distribution. We provide explicit non-asymptotic guarantees for the proposed sampler in terms of the Kullback-Leibler divergence, total variation, and Wasserstein distance to the target distribution.
• Score: 27.872857402255775
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Motivated by applications to deep learning which often fail standard Lipschitz smoothness requirements, we examine the problem of sampling from distributions that are not log-concave and are only weakly dissipative, with log-gradients allowed to grow superlinearly at infinity. In terms of structure, we only assume that the target distribution satisfies either a log-Sobolev or a Poincar\'e inequality and a local Lipschitz smoothness assumption with modulus growing possibly polynomially at infinity. This set of assumptions greatly exceeds the operational limits of the "vanilla" unadjusted Langevin algorithm (ULA), making sampling from such distributions a highly involved affair. To account for this, we introduce a taming scheme which is tailored to the growth and decay properties of the target distribution, and we provide explicit non-asymptotic guarantees for the proposed sampler in terms of the Kullback-Leibler (KL) divergence, total variation, and Wasserstein distance to the target distribution.

Related papers

• Constrained Sampling with Primal-Dual Langevin Monte Carlo [15.634831573546041]
  This work considers the problem of sampling from a probability distribution known up to a normalization constant. It satisfies a set of statistical constraints specified by the expected values of general nonlinear functions. We put forward a discrete-time primal-dual Langevin Monte Carlo algorithm (PD-LMC) that simultaneously constrains the target distribution and samples from it.
  arXiv  Detail & Related papers  (2024-11-01T13:26:13Z)
• Theory on Score-Mismatched Diffusion Models and Zero-Shot Conditional Samplers [49.97755400231656]
  We present the first performance guarantee with explicit dimensional general score-mismatched diffusion samplers. We show that score mismatches result in an distributional bias between the target and sampling distributions, proportional to the accumulated mismatch between the target and training distributions. This result can be directly applied to zero-shot conditional samplers for any conditional model, irrespective of measurement noise.
  arXiv  Detail & Related papers  (2024-10-17T16:42:12Z)
• Controllable Generation via Locally Constrained Resampling [77.48624621592523]
  We propose a tractable probabilistic approach that performs Bayesian conditioning to draw samples subject to a constraint. Our approach considers the entire sequence, leading to a more globally optimal constrained generation than current greedy methods. We show that our approach is able to steer the model's outputs away from toxic generations, outperforming similar approaches to detoxification.
  arXiv  Detail & Related papers  (2024-10-17T00:49:53Z)
• Robust Generative Learning with Lipschitz-Regularized $α$-Divergences Allows Minimal Assumptions on Target Distributions [12.19634962193403]
  This paper demonstrates the robustness of Lipschitz-regularized $alpha$-divergences as objective functionals in generative modeling. We prove the existence and finiteness of their variational derivatives, which are essential for stable training of generative models such as GANs and gradient flows. Numerical experiments confirm that generative models leveraging Lipschitz-regularized $alpha$-divergences can stably learn distributions in various challenging scenarios.
  arXiv  Detail & Related papers  (2024-05-22T19:58:13Z)
• Broadening Target Distributions for Accelerated Diffusion Models via a Novel Analysis Approach [49.97755400231656]
  We show that a novel accelerated DDPM sampler achieves accelerated performance for three broad distribution classes not considered before. Our results show an improved dependency on the data dimension $d$ among accelerated DDPM type samplers.
  arXiv  Detail & Related papers  (2024-02-21T16:11:47Z)
• Adaptive Annealed Importance Sampling with Constant Rate Progress [68.8204255655161]
  Annealed Importance Sampling (AIS) synthesizes weighted samples from an intractable distribution. We propose the Constant Rate AIS algorithm and its efficient implementation for $alpha$-divergences.
  arXiv  Detail & Related papers  (2023-06-27T08:15:28Z)
• Mean-Square Analysis of Discretized It\^o Diffusions for Heavy-tailed Sampling [17.415391025051434]
  We analyze the complexity of sampling from a class of heavy-tailed distributions by discretizing a natural class of Ito diffusions associated with weighted Poincar'e inequalities. Based on a mean-square analysis, we establish the iteration complexity for obtaining a sample whose distribution is $epsilon$ close to the target distribution in the Wasserstein-2 metric.
  arXiv  Detail & Related papers  (2023-03-01T15:16:03Z)
• Robust Estimation for Nonparametric Families via Generative Adversarial Networks [92.64483100338724]
  We provide a framework for designing Generative Adversarial Networks (GANs) to solve high dimensional robust statistics problems. Our work extend these to robust mean estimation, second moment estimation, and robust linear regression. In terms of techniques, our proposed GAN losses can be viewed as a smoothed and generalized Kolmogorov-Smirnov distance.
  arXiv  Detail & Related papers  (2022-02-02T20:11:33Z)
• Unadjusted Langevin algorithm for sampling a mixture of weakly smooth potentials [0.0]
  We prove convergence guarantees under Poincar'e inequality or non-strongly convex outside the ball. We also provide convergence in $L_beta$-Wasserstein metric for the smoothing potential.
  arXiv  Detail & Related papers  (2021-12-17T04:10:09Z)
• Distributionally Robust Bayesian Quadrature Optimization [60.383252534861136]
  We study BQO under distributional uncertainty in which the underlying probability distribution is unknown except for a limited set of its i.i.d. samples. A standard BQO approach maximizes the Monte Carlo estimate of the true expected objective given the fixed sample set. We propose a novel posterior sampling based algorithm, namely distributionally robust BQO (DRBQO) for this purpose.
  arXiv  Detail & Related papers  (2020-01-19T12:00:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.