paper-with-me

Papers

Diffusion on the Probability Simplex

2023-09-05 · Griffin Floto, Thorsteinn Jonsson, Mihai Nica, Scott Sanner, Eric Zhengyu Zhu

Diffusion models learn to reverse the progressive noising of a data distribution to create a generative model. However, the desired continuous nature of the noising process can be at odds with discrete data. To deal with this tension between continuous and discrete objects, we propose a method of performing diffusion on the probability simplex. Using the probability simplex naturally creates an interpretation where points correspond to categorical probability distributions. Our method uses the softmax function applied to an Ornstein-Unlenbeck Process, a well-known stochastic differential equation. We find that our methodology also naturally extends to include diffusion on the unit cube which has applications for bounded image generation.

📄 PDF Abstract BibTeX arXiv:2309.02530

Code (0)

등록된 구현이 없습니다.

Tasks

Image Generation

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…
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

Similar Papers 제목 키워드 기반

Categorical SDEs with Simplex Diffusion

2022-10-26 · Pierre H. Richemond, Sander Dieleman, Arnaud Doucet

Diffusion models typically operate in the standard framework of generative modelling by producing continuously-valued datapoints. To this end, they rely on a progressive Gaussian smoothing of the original data distributi…

Text Generation

Unrestrained Simplex Denoising for Discrete Data. A Non-Markovian Approach Applied to Graph Generation

2026-03-30 · Yoann Boget, Alexandros Kalousis arxiv

Denoising models such as Diffusion or Flow Matching have recently advanced generative modeling for discrete structures, yet most approaches either operate directly in the discrete state space, causing abrupt state change…

Graph Generation

Accuracy-Preserving Calibration via Statistical Modeling on Probability Simplex

2024-02-21 · Yasushi Esaki, Akihiro Nakamura, Keisuke Kawano, Ryoko Tokuhisa 외

Classification models based on deep neural networks (DNNs) must be calibrated to measure the reliability of predictions. Some recent calibration methods have employed a probabilistic model on the probability simplex. How…

Classifier calibrationUncertainty Quantification

Large-Scale Stochastic Sampling from the Probability Simplex

2018-06-19 · NeurIPS 2018 12 · Jack Baker, Paul Fearnhead, Emily B. Fox, Christopher Nemeth

Stochastic gradient Markov chain Monte Carlo (SGMCMC) has become a popular method for scalable Bayesian inference. These methods are based on sampling a discrete-time approximation to a continuous time process, such as t…

Bayesian InferenceTopic Models

Information Theoretic Bayesian Optimization over the Probability Simplex

2026-03-10 · Federico Pavesi, Antonio Candelieri, Noémie Jaquier arxiv

Bayesian optimization is a data-efficient technique that has been shown to be extremely powerful to optimize expensive, black-box, and possibly noisy objective functions. Many applications involve optimizing probabilitie…