6 papers
Conditional flow matching for physics-constrained inverse problems with finite training data
Agnimitra Dasgupta, Ali Fardisi, Mehrnegar Aminy +4
This study presents a conditional flow matching framework for solving physics-constrained Bayesian inverse problems. In this setting, samples from the joint distribution of inferre…
Closed-form conditional diffusion models for data assimilation
Brianna Binder, Agnimitra Dasgupta, Assad Oberai
We propose closed-form conditional diffusion models for data assimilation. Diffusion models use data to learn the score function (defined as the gradient of the log-probability den…
Time-dependent density estimation using binary classifiers
Agnimitra Dasgupta, Javier Murgoitio-Esandi, Ali Fardisi +1
We propose a data-driven method to learn the time-dependent probability density of a multivariate stochastic process from sample paths, assuming that the initial probability densit…
Generative Algorithms for Wildfire Progression Reconstruction from Multi-Modal Satellite Active Fire Measurements and Terrain Height
Bryan Shaddy, Brianna Binder, Agnimitra Dasgupta +8
Increasing wildfire occurrence has spurred growing interest in wildfire spread prediction. However, even the most complex wildfire models diverge from observed progression during m…
Unifying and extending Diffusion Models through PDEs for solving Inverse Problems
Agnimitra Dasgupta, Alexsander Marciano da Cunha, Ali Fardisi +4
Diffusion models have emerged as powerful generative tools with applications in computer vision and scientific machine learning (SciML), where they have been used to solve large-sc…
Memorization and Regularization in Generative Diffusion Models
Ricardo Baptista, Agnimitra Dasgupta, Nikola B. Kovachki +2
Diffusion models have emerged as a powerful framework for generative modeling. At the heart of the methodology is score matching: learning gradients of families of log-densities fo…