Sinho Chewi

activity

20212026

most citedSampling is as easy as learning the score: theory for diffusion models with minimal data assumptions

15 citations · 56 across the 38 of their papers we have counts for

collaborators

44 papers

math.ST2026

Sharp dimensional analysis of midpoint methods for Langevin sampling

Fan Chen, Sinho Chewi, Jianfeng Lu +1

We study deterministic and randomized midpoint discretizations of Langevin dynamics for a target $π \propto e^{- V}$ , where $0 ≺ α I ⪯ \nabla^{2} V ⪯ β I$ and $κ = β / α$ . To achi…

math.ST2026

High-accuracy simulation of Picard HMC, part I: Gaussian cloud correction

Fan Chen, Sinho Chewi, Jianfeng Lu +1

We study the problem of sampling from a continuous density $π \propto exp (- V)$ on $R^{d}$ , where $V \in C^{2} (R^{d})$ has a $β$ -Lipschitz gradient and $π$ satisfies a loga…

math.PR2026

Sharp Gaussian divergence and Bernstein-Markov inequalities

Fan Chen, Sinho Chewi, Jianfeng Lu +1

We prove two dimension-free estimates in Gaussian space. The first is an optimal Meyer-type inequality for the Gaussian divergence: for $p \geq 2$ , \[ \|δV\|_{L^p(γ_d)} \leq \sqrt p\…

math.ST2026

Smoothed Picard Hamiltonian Monte Carlo

Fan Chen, Sinho Chewi, Jianfeng Lu +1

We develop a new low-accuracy sampler, called smoothed Picard Hamiltonian Monte Carlo, which combines Gaussian smoothing, Picard iteration, and higher-order discretization. For a l…

math.ST2026

Accelerated High-Accuracy Sampling from a Warm Start via the Proximal Bouncy Particle Sampler

Fan Chen, Sinho Chewi, Jianfeng Lu +1

We study the problem of sampling from $μ (d x) \propto e^{- V (x)} d x$ on $R^{d}$ , where $V$ is $α$ -strongly convex and $β$ -smooth, and write $κ := β / α$ . We desi…

stat.ML2026

Generalization, memorization, and overfitting for diffusion models trained in the lazy high-dimensional regime

Hugo Latourelle-Vigeant, Sinho Chewi, Aram-Alexandre Pooladian +2

Modern score-based generative models have achieved remarkable empirical success in high-dimensional tasks such as image, audio, and video synthesis. These models reduce distributio…

cs.DS2026

Exact simulation of diffusions and improved algorithms for log-concave sampling

Fan Chen, Sinho Chewi, Alexander Rakhlin +1

We study exact simulation of diffusions via rejection sampling on path space using unbiased estimators of the density ratio obtained from Girsanov's theorem. When applied to the un…

math.PR2026

Near-Lipschitz stability of the Kim--Milman flow map

Sinho Chewi, Katharina Eichinger, Aram-Alexandre Pooladian

We prove that the Kim--Milman flow map enjoys favorable stability properties with respect to variations in the target measure, provided that one of the target measures is sufficien…

cs.DS2026

Complexity of Non-Log-Concave Sampling in Fisher Information

Sinho Chewi, Andre Wibisono

We study the query complexity of obtaining a relative Fisher information guarantee for sampling from a log-smooth non-log-concave distribution; this is a sampling analog of finding…

math.ST2026

A proximal gradient algorithm for composite log-concave sampling

Linghai Liu, Sinho Chewi

We propose an algorithm to sample from composite log-concave distributions over $R^{d}$ , i.e., densities of the form $π \propto e^{- f - g}$ , assuming access to gradient evaluat…

math.OC2026

Lectures on optimization

Sinho Chewi

These lecture notes cover the theory of convex optimization, with a particular emphasis on first-order methods.

cs.LG2026

A Rod Flow Model for Adam at the Edge of Stability

Eric Regis, Sinho Chewi

Cohen et al. (arXiv:2207.14484) observed that adaptive gradient methods such as Adam operate at the edge of stability. While there has been significant work on continuous-time mode…

cs.LG2026

Variational inference via radial transport

Luca Ghafourpour, Sinho Chewi, Alessio Figalli +1

In variational inference (VI), the practitioner approximates a high-dimensional distribution $π$ with a simple surrogate one, often a (product) Gaussian distribution. However, in m…

cs.DS2026

Algorithmic warm starts for Hamiltonian Monte Carlo

Matthew S. Zhang, Jason M. Altschuler, Sinho Chewi

Generating samples from a continuous probability density is a central algorithmic problem across statistics, engineering, and the sciences. For high-dimensional settings, Hamiltoni…

math.ST2026

Sampling from Constrained Gibbs Measures: with Applications to High-Dimensional Bayesian Inference

Ruixiao Wang, Xiaohong Chen, Sinho Chewi

This paper considers a non-standard problem of generating samples from a low-temperature Gibbs distribution with \emph{constrained} support, when some of the coordinates of the mod…

math.ST2026

High-accuracy log-concave sampling with stochastic queries

Fan Chen, Sinho Chewi, Constantinos Daskalakis +1

We show that high-accuracy guarantees for log-concave sampling -- that is, iteration and query complexities which scale as $poly lo g (1/ δ)$ , where $δ$ is the desired target…

cs.LG2026

Blind denoising diffusion models and the blessings of dimensionality

Zahra Kadkhodaie, Aram-Alexandre Pooladian, Sinho Chewi +1

Denoising diffusion models (DDMs) are state-of-the-art methods for learning densities from data across numerous domains, yet many aspects of the training and sampling pipeline rema…

cs.LG2026

Rod Flow: A Continuous-Time Model for Gradient Descent at the Edge of Stability

Eric Regis, Sinho Chewi

How can we understand gradient-based training over non-convex landscapes? The edge of stability phenomenon, introduced in Cohen et al. (2021), indicates that the answer is not so s…

cs.LG2026

High-accuracy sampling for diffusion models and log-concave distributions

Fan Chen, Sinho Chewi, Constantinos Daskalakis +1

We present algorithms for diffusion model sampling which obtain $δ$ -error in $polylog (1/ δ)$ steps, given access to $O (δ)$ -accurate score estimates in $L^{2}$ . Thi…

stat.ML2025

Theory and computation for structured variational inference

Shunan Sheng, Bohan Wu, Bennett Zhu +2

Structured variational inference constitutes a core methodology in modern statistical applications. Unlike mean-field variational inference, the approximate posterior is assumed to…

cs.LG2025

Sublinear iterations can suffice even for DDPMs

Matthew S. Zhang, Stephen Huan, Jerry Huang +3

SDE-based methods such as denoising diffusion probabilistic models (DDPMs) have shown remarkable success in real-world sample generation tasks. Prior analyses of DDPMs have been fo…

math.PR2025

Stability of the Kim--Milman flow map

Sinho Chewi, Aram-Alexandre Pooladian, Matthew S. Zhang

In this short note, we characterize stability of the Kim--Milman flow map -- also known as the probability flow ODE -- with respect to variations in the target measure in relative…

math.PR2025

Shifted Composition IV: Toward Ballistic Acceleration for Log-Concave Sampling

Jason M. Altschuler, Sinho Chewi, Matthew S. Zhang

Acceleration is a celebrated cornerstone of convex optimization, enabling gradient-based algorithms to converge sublinearly in the condition number. A major open question is whethe…

math.ST2024

Shifted Composition III: Local Error Framework for KL Divergence

Jason M. Altschuler, Sinho Chewi

Coupling arguments are a central tool for bounding the deviation between two stochastic processes, but traditionally have been limited to Wasserstein metrics. In this paper, we app…