English
Related papers

Related papers: KL Convergence Guarantees for Score diffusion mode…

200 papers

We compute the expected value of the Kullback-Leibler divergence to various fundamental statistical models with respect to canonical priors on the probability simplex. We obtain closed formulas for the expected model approximation errors,…

Machine Learning · Statistics 2014-06-18 Guido F. Montufar , Johannes Rauh

We investigate the statistical and computational limits of latent Diffusion Transformers (DiTs) under the low-dimensional linear latent space assumption. Statistically, we study the universal approximation and sample complexity of the DiTs…

Machine Learning · Statistics 2024-11-01 Jerry Yao-Chieh Hu , Weimin Wu , Zhao Song , Han Liu

We interpret likelihood-based test functions from a geometric perspective where the Kullback-Leibler (KL) divergence is adopted to quantify the distance from a distribution to another. Such a test function can be seen as a sub-Gaussian…

Information Theory · Computer Science 2021-01-05 Yan Wang

This work introduces a sampling method capable of solving Bayesian inverse problems in function space. It does not assume the log-concavity of the likelihood, meaning that it is compatible with nonlinear inverse problems. The method…

Machine Learning · Statistics 2024-05-27 Lorenzo Baldassari , Ali Siahkoohi , Josselin Garnier , Knut Solna , Maarten V. de Hoop

While continuous diffusion models excel in modeling continuous distributions, their application to categorical data has been less effective. Recent work has shown that ratio-matching through score-entropy within a continuous-time discrete…

Machine Learning · Statistics 2026-02-09 Etrit Haxholli , Yeti Z. Gurbuz , Ogul Can , Eli Waxman

In a first part, we present a mathematical analysis of a general methodology of a probabilistic learning inference that allows for estimating a posterior probability model for a stochastic boundary value problem from a prior probability…

Machine Learning · Statistics 2022-06-08 Christian Soize

Generative diffusion models are extensively used in unsupervised and self-supervised machine learning with the aim to generate new samples from a probability distribution estimated with a set of known samples. They have demonstrated…

Fluid Dynamics · Physics 2026-01-28 Wilfried Genuist , Éric Savin , Filippo Gatti , Didier Clouteau

In this paper, we investigate the convergence of language models (LMs) trained under different random seeds, measuring convergence as the expected per-token Kullback--Leibler (KL) divergence across seeds. By comparing LM convergence as a…

Computation and Language · Computer Science 2025-10-01 Finlay Fehlauer , Kyle Mahowald , Tiago Pimentel

In Learning Theory, the smoothness assumption on the target function (known as source condition) is a key factor in establishing theoretical convergence rates for an estimator. The existing general form of the source condition, as discussed…

Statistics Theory · Mathematics 2025-03-27 Naveen Gupta , S. Sivananthan

Score-based diffusion models are a highly effective method for generating samples from a distribution of images. We consider scenarios where the training data comes from a noisy version of the target distribution, and present an efficiently…

Machine Learning · Statistics 2025-09-30 Dennis Elbrächter , Giovanni S. Alberti , Matteo Santacesaria

We investigate the approximation efficiency of score functions by deep neural networks in diffusion-based generative modeling. While existing approximation theories utilize the smoothness of score functions, they suffer from the curse of…

Machine Learning · Computer Science 2023-09-21 Song Mei , Yuchen Wu

Recently, a series of papers proposed deep learning-based approaches to sample from target distributions using controlled diffusion processes, being trained only on the unnormalized target densities without access to samples. Building on…

Machine Learning · Computer Science 2024-05-24 Lorenz Richter , Julius Berner

We construct and analyze generative diffusions that transport a point mass to a prescribed target distribution over a finite time horizon using the stochastic interpolant framework. The drift is expressed as a conditional expectation that…

Statistics Theory · Mathematics 2026-05-21 Yifan Chen , Eric Vanden-Eijnden

Decision-focused learning (DFL) integrates predictive modeling and optimization by training predictors to optimize the downstream decision target rather than merely minimizing prediction error. To date, existing DFL methods typically rely…

Machine Learning · Computer Science 2025-10-14 Zihao Zhao , Christopher Yeh , Lingkai Kong , Kai Wang

Diffusion models have emerged as the principal paradigm for generative modeling across various domains. During training, they learn the score function, which in turn is used to generate samples at inference. They raise a basic yet unsolved…

Machine Learning · Computer Science 2025-10-03 Kiwhan Song , Jaeyeon Kim , Sitan Chen , Yilun Du , Sham Kakade , Vincent Sitzmann

Score-based diffusion models have emerged as one of the most promising frameworks for deep generative modelling, due to their state-of-the art performance in many generation tasks while relying on mathematical foundations such as stochastic…

Machine Learning · Computer Science 2023-11-28 Teo Deveney , Jan Stanczuk , Lisa Maria Kreusser , Chris Budd , Carola-Bibiane Schönlieb

This paper introduces Discrete Markov Probabilistic Models (DMPMs), a novel discrete diffusion algorithm for discrete data generation. The algorithm operates in discrete bit space, where the noising process is a continuous-time Markov chain…

Machine Learning · Statistics 2025-10-09 Le-Tuyet-Nhi Pham , Dario Shariatian , Antonio Ocello , Giovanni Conforti , Alain Durmus

Score-based diffusion models have emerged as one of the most promising frameworks for deep generative modelling. In this work we conduct a systematic comparison and theoretical analysis of different approaches to learning conditional…

Machine Learning · Computer Science 2021-11-29 Georgios Batzolis , Jan Stanczuk , Carola-Bibiane Schönlieb , Christian Etmann

Expectation maximization (EM) is the default algorithm for fitting probabilistic models with missing or latent variables, yet we lack a full understanding of its non-asymptotic convergence properties. Previous works show results along the…

Machine Learning · Computer Science 2022-03-01 Frederik Kunstner , Raunak Kumar , Mark Schmidt

We give a new algorithm for learning mixtures of $k$ Gaussians (with identity covariance in $\mathbb{R}^n$) to TV error $\varepsilon$, with quasi-polynomial ($O(n^{\text{poly\,log}\left(\frac{n+k}{\varepsilon}\right)})$) time and sample…

Machine Learning · Computer Science 2025-03-05 Khashayar Gatmiry , Jonathan Kelner , Holden Lee