Výsledky vyhledávání - "Patel, Dhruv V"

Report

Autor: Patel, Dhruv V., Lee, Jonghyun, Farthing, Matthew W., Kitanidis, Peter K., Darve, Eric F.

Numerous applications in biology, statistics, science, and engineering require generating samples from high-dimensional probability distributions. In recent years, the Hamiltonian Monte Carlo (HMC) method has emerged as a state-of-the-art Markov chai

Externí odkaz: http://arxiv.org/abs/2405.05033

Zobrazit plný text záznamu

Report

A dimension-reduced variational approach for solving physics-based inverse problems using generative adversarial network priors and normalizing flows

Autor: Dasgupta, Agnimitra, Patel, Dhruv V, Ray, Deep, Johnson, Erik A, Oberai, Assad A

We propose a novel modular inference approach combining two different generative models -- generative adversarial networks (GAN) and normalizing flows -- to approximate the posterior distribution of physics-based Bayesian inverse problems framed in h

Externí odkaz: http://arxiv.org/abs/2310.04690

Zobrazit plný text záznamu

Report

The efficacy and generalizability of conditional GANs for posterior inference in physics-based inverse problems

Autor: Ray, Deep, Ramaswamy, Harisankar, Patel, Dhruv V., Oberai, Assad A.

In this work, we train conditional Wasserstein generative adversarial networks to effectively sample from the posterior of physics-based Bayesian inference problems. The generator is constructed using a U-Net architecture, with the latent information

Externí odkaz: http://arxiv.org/abs/2202.07773

Zobrazit plný text záznamu

Report

Solution of Physics-based Bayesian Inverse Problems with Deep Generative Priors

Autor: Patel, Dhruv V, Ray, Deep, Oberai, Assad A

Inverse problems are ubiquitous in nature, arising in almost all areas of science and engineering ranging from geophysics and climate science to astrophysics and biomechanics. One of the central challenges in solving inverse problems is tackling thei

Externí odkaz: http://arxiv.org/abs/2107.02926

Zobrazit plný text záznamu

Akademický článek

A dimension-reduced variational approach for solving physics-based inverse problems using generative adversarial network priors and normalizing flows

Autor: Dasgupta, Agnimitra, Patel, Dhruv V., Ray, Deep, Johnson, Erik A., Oberai, Assad A.

Publikováno v: In Computer Methods in Applied Mechanics and Engineering 15 February 2024 420

Zobrazit plný text záznamu

Report

GAN-based Priors for Quantifying Uncertainty

Autor: Patel, Dhruv V., Oberai, Assad A.

Bayesian inference is used extensively to quantify the uncertainty in an inferred field given the measurement of a related field when the two are linked by a mathematical model. Despite its many applications, Bayesian inference faces challenges when

Externí odkaz: http://arxiv.org/abs/2003.12597

Zobrazit plný text záznamu