1 code implementation • 15 Mar 2024 • Declan McNamara, Jackson Loper, Jeffrey Regier
As an alternative, we propose SMC-Wake, a procedure for fitting an amortized variational approximation that uses likelihood-tempered sequential Monte Carlo samplers to estimate the gradient of the inclusive KL divergence.
1 code implementation • 20 Jul 2023 • Zhiwei Xue, Yuhang Li, Yash Patel, Jeffrey Regier
As an alternative, we propose a classifier-free conditional diffusion model for PSF deconvolution of galaxy images.
no code implementations • 23 May 2023 • Yash Patel, Declan McNamara, Jackson Loper, Jeffrey Regier, Ambuj Tewari
We prove lower bounds on the predictive efficiency of the regions produced by CANVI and explore how the quality of a posterior approximation relates to the predictive efficiency of prediction regions based on that approximation.
1 code implementation • 17 Nov 2022 • Mallory Wang, Ismael Mendoza, Cheng Wang, Camille Avestruz, Jeffrey Regier
Coadded astronomical images are created by stacking multiple single-exposure images.
no code implementations • 25 Oct 2022 • Prayag Chatha, Yixin Wang, Zhenke Wu, Jeffrey Regier
In medicine, researchers often seek to infer the effects of a given treatment on patients' outcomes.
1 code implementation • 12 Jul 2022 • Derek Hansen, Ismael Mendoza, Runjing Liu, Ziteng Pang, Zhe Zhao, Camille Avestruz, Jeffrey Regier
We present a new probabilistic method for detecting, deblending, and cataloging astronomical sources called the Bayesian Light Source Separator (BLISS).
1 code implementation • 3 Jun 2021 • Derek Hansen, Brian Manzo, Jeffrey Regier
We propose a new procedure called FlowSelect to perform controlled feature selection that does not suffer from either of these two problems.
3 code implementations • 4 Feb 2021 • Runjing Liu, Jon D. McAuliffe, Jeffrey Regier
In images collected by astronomical surveys, stars and galaxies often overlap visually.
1 code implementation • 17 Jun 2020 • Tianci Liu, Jeffrey Regier
Generative adversarial networks (GANs) and normalizing flows are both approaches to density estimation that use deep neural networks to transform samples from an uninformative prior distribution to an approximation of the data distribution.
2 code implementations • NeurIPS 2020 • Romain Lopez, Pierre Boyeau, Nir Yosef, Michael. I. Jordan, Jeffrey Regier
To make decisions based on a model fit with auto-encoding variational Bayes (AEVB), practitioners often let the variational distribution serve as a surrogate for the posterior distribution.
2 code implementations • 6 May 2019 • Romain Lopez, Achille Nazaret, Maxime Langevin, Jules Samaran, Jeffrey Regier, Michael. I. Jordan, Nir Yosef
Building upon domain adaptation work, we propose gimVI, a deep generative model for the integration of spatial transcriptomic data and scRNA-seq data that can be used to impute missing genes.
1 code implementation • 10 Oct 2018 • Runjing Liu, Jeffrey Regier, Nilesh Tripuraneni, Michael. I. Jordan, Jon McAuliffe
We wish to compute the gradient of an expectation over a finite or countably infinite sample space having $K \leq \infty$ categories.
1 code implementation • 16 Sep 2018 • Maxime Langevin, Edouard Mehlman, Jeffrey Regier, Romain Lopez, Michael. I. Jordan, Nir Yosef
Class labels are often imperfectly observed, due to mistakes and to genuine ambiguity among classes.
no code implementations • NeurIPS 2018 • Romain Lopez, Jeffrey Regier, Michael. I. Jordan, Nir Yosef
We show how to apply this method to a range of problems, including the problems of learning invariant representations and the learning of interpretable representations.
1 code implementation • 28 Feb 2018 • Jeffrey Regier, Andrew C. Miller, David Schlegel, Ryan P. Adams, Jon D. McAuliffe, Prabhat
We present a new, fully generative model for constructing astronomical catalogs from optical telescope image sets.
1 code implementation • 31 Jan 2018 • Jeffrey Regier, Kiran Pamnany, Keno Fischer, Andreas Noack, Maximilian Lam, Jarrett Revels, Steve Howard, Ryan Giordano, David Schlegel, Jon McAuliffe, Rollin Thomas, Prabhat
We construct an astronomical catalog from 55 TB of imaging data using Celeste, a Bayesian variational inference code written entirely in the high-productivity programming language Julia.
Distributed, Parallel, and Cluster Computing Instrumentation and Methods for Astrophysics 85A35, 68W10, 62P35 J.2; D.1.3; G.3; I.2; D.2
no code implementations • NeurIPS 2018 • Nilesh Tripuraneni, Mitchell Stern, Chi Jin, Jeffrey Regier, Michael. I. Jordan
This paper proposes a stochastic variant of a classic algorithm---the cubic-regularized Newton method [Nesterov and Polyak 2006].
no code implementations • 13 Oct 2017 • Romain Lopez, Jeffrey Regier, Michael Cole, Michael Jordan, Nir Yosef
We propose a probabilistic model for interpreting gene expression levels that are observed through single-cell RNA sequencing.
2 code implementations • 7 Sep 2017 • Romain Lopez, Jeffrey Regier, Michael Cole, Michael Jordan, Nir Yosef
We also extend our framework to account for batch effects and other confounding factors, and propose a Bayesian hypothesis test for differential expression that outperforms DESeq2.
no code implementations • NeurIPS 2017 • Jeffrey Regier, Michael. I. Jordan, Jon McAuliffe
We introduce TrustVI, a fast second-order algorithm for black-box variational inference based on trust-region optimization and the reparameterization trick.
no code implementations • 10 Nov 2016 • Jeffrey Regier, Kiran Pamnany, Ryan Giordano, Rollin Thomas, David Schlegel, Jon McAuliffe, Prabhat
Celeste is a procedure for inferring astronomical catalogs that attains state-of-the-art scientific results.
no code implementations • 3 Jun 2015 • Jeffrey Regier, Andrew Miller, Jon McAuliffe, Ryan Adams, Matt Hoffman, Dustin Lang, David Schlegel, Prabhat
We present a new, fully generative model of optical telescope image sets, along with a variational procedure for inference.