32 citations · 74 across the 17 of their papers we have counts for
17 papers
Gaussian mixture layers for neural networks
Sinho Chewi, Philippe Rigollet, Yuling Yan
The mean-field theory for two-layer neural networks considers infinitely wide networks that are linearly parameterized by a probability measure over the parameter space. This nonpa…
DDPM Score Matching and Distribution Learning
Sinho Chewi, Alkis Kalavasis, Anay Mehrotra +1
Score estimation is the backbone of score-based generative models (SGMs), especially denoising diffusion probabilistic models (DDPMs). A key result in this area shows that with acc…
The ballistic limit of the log-Sobolev constant equals the Polyak-Łojasiewicz constant
Sinho Chewi, Austin J. Stromme
The Polyak-Lojasiewicz (PL) constant of a function characterizes the best exponential rate of convergence of gradient flow for , uniformly…
Statistical optimal transport
Sinho Chewi, Jonathan Niles-Weed, Philippe Rigollet
We present an introduction to the field of statistical optimal transport, based on lectures given at École d'Été de Probabilités de Saint-Flour XLIX.
Learning threshold neurons via the "edge of stability"
Kwangjun Ahn, Sébastien Bubeck, Sinho Chewi +3
Existing analyses of neural network training often operate under the unrealistic assumption of an extremely small learning rate. This lies in stark contrast to practical wisdom and…
The entropic barrier is -self-concordant
Sinho Chewi
For any convex body , S. Bubeck and R. Eldan introduced the entropic barrier on and showed that it is a -self-concordant barrier. In thi…
The query complexity of sampling from strongly log-concave distributions in one dimension
Sinho Chewi, Patrik Gerber, Chen Lu +2
We establish the first tight lower bound of on the query complexity of sampling from the class of strongly log-concave and log-smooth distributions with condition nu…
Rejection sampling from shape-constrained distributions in sublinear time
Sinho Chewi, Patrik Gerber, Chen Lu +2
We consider the task of generating exact samples from a target distribution, known up to normalization, over a finite alphabet. The classical algorithm for this task is rejection s…
Dimension-free log-Sobolev inequalities for mixture distributions
Hong-Bin Chen, Sinho Chewi, Jonathan Niles-Weed
We prove that if is a family of probability measures which satisfy the log-Sobolev inequality and whose pairwise chi-squared divergences are uniformly b…
Optimal dimension dependence of the Metropolis-Adjusted Langevin Algorithm
Sinho Chewi, Chen Lu, Kwangjun Ahn +3
Conventional wisdom in the sampling literature, backed by a popular diffusion scaling limit, suggests that the mixing time of the Metropolis-Adjusted Langevin Algorithm (MALA) scal…
Fast and Smooth Interpolation on Wasserstein Space
Sinho Chewi, Julien Clancy, Thibaut Le Gouic +3
We propose a new method for smoothly interpolating probability measures using the geometry of optimal transport. To that end, we reduce this problem to the classical Euclidean sett…
Efficient constrained sampling via the mirror-Langevin algorithm
Kwangjun Ahn, Sinho Chewi
We propose a new discretization of the mirror-Langevin diffusion and give a crisp proof of its convergence. Our analysis uses relative convexity/smoothness and self-concordance, id…
SVGD as a kernelized Wasserstein gradient flow of the chi-squared divergence
Sinho Chewi, Thibaut Le Gouic, Chen Lu +2
Stein Variational Gradient Descent (SVGD), a popular sampling algorithm, is often described as the kernelized gradient flow for the Kullback-Leibler divergence in the geometry of o…
Exponential ergodicity of mirror-Langevin diffusions
Sinho Chewi, Thibaut Le Gouic, Chen Lu +3
Motivated by the problem of sampling from ill-conditioned log-concave distributions, we give a clean non-asymptotic convergence analysis of mirror-Langevin diffusions as introduced…
Gradient descent algorithms for Bures-Wasserstein barycenters
Sinho Chewi, Tyler Maunu, Philippe Rigollet +1
We study first order methods to compute the barycenter of a probability distribution over the space of probability measures with finite second moment. We develop a framework to…
Matching Observations to Distributions: Efficient Estimation via Sparsified Hungarian Algorithm
Sinho Chewi, Forest Yang, Avishek Ghosh +2
Suppose we are given observations, where each observation is drawn independently from one of known distributions. The goal is to match each observation to the distribution from…
A combinatorial proof of a formula of Biane and Chapuy
Sinho Chewi, Venkat Anantharam
Let be a simple strongly connected weighted directed graph. Let denote the spanning tree graph of . That is, the vertices of consist of the direc…