Score-based diffusion models in function space
Diffusion models have recently emerged as a powerful framework for generative modeling.
They consist of a forward process that perturbs input data with Gaussian white noise and a …
They consist of a forward process that perturbs input data with Gaussian white noise and a …
Approximation and optimization theory for linear continuous-time recurrent neural networks
We perform a systematic study of the approximation properties and optimization dynamics of
recurrent neural networks (RNNs) when applied to learn input-output relationships in …
recurrent neural networks (RNNs) when applied to learn input-output relationships in …
Convergence of Stein variational gradient descent under a weaker smoothness condition
L Sun, A Karagulyan… - … Conference on Artificial …, 2023 - proceedings.mlr.press
Abstract Stein Variational Gradient Descent (SVGD) is an important alternative to the
Langevin-type algorithms for sampling from probability distributions of the form $\pi …
Langevin-type algorithms for sampling from probability distributions of the form $\pi …
Online lewis weight sampling
DP Woodruff, T Yasuda - Proceedings of the 2023 Annual ACM-SIAM …, 2023 - SIAM
The seminal work of Cohen and Peng [CP15](STOC 2015) introduced Lewis weight
sampling to the theoretical computer science community, which yields fast row sampling …
sampling to the theoretical computer science community, which yields fast row sampling …
Some properties of univariate and multivariate exponential power distributions and related topics
V Korolev - Mathematics, 2020 - mdpi.com
In the paper, a survey of the main results concerning univariate and multivariate exponential
power (EP) distributions is given, with main attention paid to mixture representations of these …
power (EP) distributions is given, with main attention paid to mixture representations of these …
On closed-form expressions for the Fisher-Rao distance
The Fisher-Rao distance is the geodesic distance between probability distributions in a
statistical manifold equipped with the Fisher metric, which is a natural choice of Riemannian …
statistical manifold equipped with the Fisher metric, which is a natural choice of Riemannian …
Distributed detection of sparse stochastic signals with quantized measurements: The generalized Gaussian case
In this paper, we consider distributed detection of sparse stochastic signals with quantized
measurements. Assume that both the noise and the dominant elements in sparse signals …
measurements. Assume that both the noise and the dominant elements in sparse signals …
Robust neural networks using randomized adversarial training
This paper tackles the problem of defending a neural network against adversarial attacks
crafted with different norms (in particular $\ell_\infty $ and $\ell_2 $ bounded adversarial …
crafted with different norms (in particular $\ell_\infty $ and $\ell_2 $ bounded adversarial …
p-Generalized probit regression and scalable maximum likelihood estimation via sketching and coresets
A Munteanu, S Omlor, C Peters - … Conference on Artificial …, 2022 - proceedings.mlr.press
We study the $ p $-generalized probit regression model, which is a generalized linear model
for binary responses. It extends the standard probit model by replacing its link function, the …
for binary responses. It extends the standard probit model by replacing its link function, the …
New sparsity measure based on energy distribution
E Mahmoudian, H Amindavar, SM Ahadi - Displays, 2023 - Elsevier
Sparsity has long been a theoretical and practical signal property in applied mathematics
and is utilized as a crucial concept in signal/image processing applications such as …
and is utilized as a crucial concept in signal/image processing applications such as …