paper-with-me

홈 › Papers

Learning Mixtures of Gaussians Using the DDPM Objective

2023-07-03 · NeurIPS 2023 11

Recent works have shown that diffusion models can learn essentially any distribution provided one can perform score estimation. Yet it remains poorly understood under what settings score estimation is possible, let alone when practical gradient-based algorithms for this task can provably succeed. In this work, we give the first provably efficient results along these lines for one of the most fundamental distribution families, Gaussian mixture models. We prove that gradient descent on the denoising diffusion probabilistic model (DDPM) objective can efficiently recover the ground truth parameters of the mixture model in the following two settings: 1) We show gradient descent with random initialization learns mixtures of two spherical Gaussians in $d$ dimensions with $1/\text{poly}(d)$-separated centers. 2) We show gradient descent with a warm start learns mixtures of $K$ spherical Gaussians with $\Omega(\sqrt{\log(\min(K,d))})$-separated centers. A key ingredient in our proofs is a new connection between score-based methods and two other approaches to distribution learning, the EM algorithm and spectral methods.

📄 PDF Abstract BibTeX arXiv:2307.01178

Code (0)

등록된 구현이 없습니다.

Tasks

Denoising

Methods 이 논문이 사용한 방법론

Diffusion Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound…

Similar Papers 제목 키워드 기반

Sample-Efficient Private Learning of Mixtures of Gaussians

2024-11-04 · Hassan Ashtiani, Mahbod Majid, Shyam Narayanan

We study the problem of learning mixtures of Gaussians with approximate differential privacy. We prove that roughly $kd^2 + k^{1.5} d^{1.75} + k^2 d$ samples suffice to learn a mixture of $k$ arbitrary $d$-dimensional Ga…

Privately Learning Mixtures of Axis-Aligned Gaussians

2021-06-03 · NeurIPS 2021 12 · Ishaq Aden-Ali, Hassan Ashtiani, Christopher Liaw

We consider the problem of learning mixtures of Gaussians under the constraint of approximate differential privacy. We prove that $\widetilde{O}(k^2 d \log^{3/2}(1/\delta) / \alpha^2 \varepsilon)$ samples are sufficient …

Mixtures of Gaussians are Privately Learnable with a Polynomial Number of Samples

2023-09-07 · Mohammad Afzali, Hassan Ashtiani, Christopher Liaw

We study the problem of estimating mixtures of Gaussians under the constraint of differential privacy (DP). Our main result is that $\text{poly}(k,d,1/\alpha,1/\varepsilon,\log(1/\delta))$ samples are sufficient to estim…

Learning Mixtures of Gaussians with Censored Data

2023-05-06 · Wai Ming Tai, Bryon Aragam

We study the problem of learning mixtures of Gaussians with censored data. Statistical learning with censored data is a classical problem, with numerous practical applications, however, finite-sample guarantees for even …

Nearly tight sample complexity bounds for learning mixtures of Gaussians via sample compression schemes

2018-12-01 · NeurIPS 2018 12 · Hassan Ashtiani, Shai Ben-David, Nicholas Harvey, Christopher Liaw 외

We prove that ϴ(k d^2 / ε^2) samples are necessary and sufficient for learning a mixture of k Gaussians in R^d, up to error ε in total variation distance. This improves both the known upper bounds and lower bounds for th…