From Score Matching to Diffusion: A Fine-Grained Error Analysis in the Gaussian Setting

• URL: http://arxiv.org/abs/2503.11615v2
• Date: Fri, 23 May 2025 12:56:55 GMT
• Title: From Score Matching to Diffusion: A Fine-Grained Error Analysis in the Gaussian Setting
• Authors: Samuel Hurault, Matthieu Terris, Thomas Moreau, Gabriel Peyré,
• Abstract summary: We show that the Wasserstein sampling error can be expressed as a kernel-type norm of the data power spectrum.<n>We show that the Wasserstein sampling error can be expressed as a kernel-type norm of the data power spectrum.
• Score: 25.21429354164613
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Sampling from an unknown distribution, accessible only through discrete samples, is a fundamental problem at the core of generative AI. The current state-of-the-art methods follow a two-step process: first, estimating the score function (the gradient of a smoothed log-distribution) and then applying a diffusion-based sampling algorithm -- such as Langevin or Diffusion models. The resulting distribution's correctness can be impacted by four major factors: the generalization and optimization errors in score matching, and the discretization and minimal noise amplitude in the diffusion. In this paper, we make the sampling error explicit when using a diffusion sampler in the Gaussian setting. We provide a sharp analysis of the Wasserstein sampling error that arises from these four error sources. This allows us to rigorously track how the anisotropy of the data distribution (encoded by its power spectrum) interacts with key parameters of the end-to-end sampling method, including the number of initial samples, the stepsizes in both score matching and diffusion, and the noise amplitude. Notably, we show that the Wasserstein sampling error can be expressed as a kernel-type norm of the data power spectrum, where the specific kernel depends on the method parameters. This result provides a foundation for further analysis of the tradeoffs involved in optimizing sampling accuracy.

Related papers

• Convergence of Deterministic and Stochastic Diffusion-Model Samplers: A Simple Analysis in Wasserstein Distance [0.0]
  We provide convergence guarantees in Wasserstein distance for diffusion-based generative models, covering both (DDPM-like) and deterministic (DDIM-like) sampling methods.<n> Notably, we derive the first Wasserstein convergence bound for the Heun sampler and improve existing results for the sampler of the probability flow ODE.
  arXiv  Detail & Related papers  (2025-08-05T08:37:58Z)
• Sampling through Algorithmic Diffusion in non-convex Perceptron problems [2.860608352191896]
  We introduce a formalism based on the replica method to characterize the feasibility in terms of a few order parameters.<n>We show that, in the case of the perceptron problem with negative stability, approximate uniform sampling is achievable across the entire replica region.<n>In contrast, for the binary perceptron, uniform sampling via diffusion invariably fails due to the overlap gap property exhibited by the typical set of solutions.
  arXiv  Detail & Related papers  (2025-02-22T16:43:01Z)
• Distributional Diffusion Models with Scoring Rules [83.38210785728994]
  Diffusion models generate high-quality synthetic data.<n> generating high-quality outputs requires many discretization steps.<n>We propose to accomplish sample generation by learning the posterior em distribution of clean data samples.
  arXiv  Detail & Related papers  (2025-02-04T16:59:03Z)
• On the Wasserstein Convergence and Straightness of Rectified Flow [54.580605276017096]
  Rectified Flow (RF) is a generative model that aims to learn straight flow trajectories from noise to data. We provide a theoretical analysis of the Wasserstein distance between the sampling distribution of RF and the target distribution. We present general conditions guaranteeing uniqueness and straightness of 1-RF, which is in line with previous empirical findings.
  arXiv  Detail & Related papers  (2024-10-19T02:36:11Z)
• Unified Convergence Analysis for Score-Based Diffusion Models with Deterministic Samplers [49.1574468325115]
  We introduce a unified convergence analysis framework for deterministic samplers. Our framework achieves iteration complexity of $tilde O(d2/epsilon)$. We also provide a detailed analysis of Denoising Implicit Diffusion Models (DDIM)-type samplers.
  arXiv  Detail & Related papers  (2024-10-18T07:37:36Z)
• Theory on Score-Mismatched Diffusion Models and Zero-Shot Conditional Samplers [49.97755400231656]
  We present the first performance guarantee with explicit dimensional dependencies for general score-mismatched diffusion samplers.<n>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.<n>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)
• Noise-Free Sampling Algorithms via Regularized Wasserstein Proximals [3.4240632942024685]
  We consider the problem of sampling from a distribution governed by a potential function. This work proposes an explicit score based MCMC method that is deterministic, resulting in a deterministic evolution for particles.
  arXiv  Detail & Related papers  (2023-08-28T23:51:33Z)
• Bayesian Analysis for Over-parameterized Linear Model via Effective Spectra [6.9060054915724]
  We introduce a data-adaptive Gaussian prior that targets the data's intrinsic complexity rather than its ambient dimension.<n>We establish contraction rates of the corresponding posterior distribution, which reveal how the mass in the spectrum affects the prediction error bounds.<n>Our findings demonstrate that Bayesian methods leveraging spectral information of the data are effective for estimation in non-sparse, high-dimensional settings.
  arXiv  Detail & Related papers  (2023-05-25T06:07:47Z)
• Restoration-Degradation Beyond Linear Diffusions: A Non-Asymptotic Analysis For DDIM-Type Samplers [90.45898746733397]
  We develop a framework for non-asymptotic analysis of deterministic samplers used for diffusion generative modeling. We show that one step along the probability flow ODE can be expressed as two steps: 1) a restoration step that runs ascent on the conditional log-likelihood at some infinitesimally previous time, and 2) a degradation step that runs the forward process using noise pointing back towards the current gradient.
  arXiv  Detail & Related papers  (2023-03-06T18:59:19Z)
• Unsupervised Learning of Sampling Distributions for Particle Filters [80.6716888175925]
  We put forward four methods for learning sampling distributions from observed measurements. Experiments demonstrate that learned sampling distributions exhibit better performance than designed, minimum-degeneracy sampling distributions.
  arXiv  Detail & Related papers  (2023-02-02T15:50:21Z)
• Unrolling Particles: Unsupervised Learning of Sampling Distributions [102.72972137287728]
  Particle filtering is used to compute good nonlinear estimates of complex systems. We show in simulations that the resulting particle filter yields good estimates in a wide range of scenarios.
  arXiv  Detail & Related papers  (2021-10-06T16:58:34Z)
• SNIPS: Solving Noisy Inverse Problems Stochastically [25.567566997688044]
  We introduce a novel algorithm dubbed SNIPS, which draws samples from the posterior distribution of any linear inverse problem. Our solution incorporates ideas from Langevin dynamics and Newton's method, and exploits a pre-trained minimum mean squared error (MMSE) We show that the samples produced are sharp, detailed and consistent with the given measurements, and their diversity exposes the inherent uncertainty in the inverse problem being solved.
  arXiv  Detail & Related papers  (2021-05-31T13:33:21Z)
• Learning Entangled Single-Sample Distributions via Iterative Trimming [28.839136703139225]
  We analyze a simple and computationally efficient method based on iteratively trimming samples and re-estimating the parameter on the trimmed sample set. We show that the method in logarithmic iterations outputs an estimation whose error only depends on the noise level of the $lceil alpha n rceil$-th noisiest data point.
  arXiv  Detail & Related papers  (2020-04-20T18:37:43Z)
• Generative Modeling with Denoising Auto-Encoders and Langevin Sampling [88.83704353627554]
  We show that both DAE and DSM provide estimates of the score of the smoothed population density. We then apply our results to the homotopy method of arXiv:1907.05600 and provide theoretical justification for its empirical success.
  arXiv  Detail & Related papers  (2020-01-31T23:50:03Z)

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.