Posterior Approximation

Posterior approximation focuses on efficiently estimating complex probability distributions, crucial for Bayesian inference in diverse fields. Current research emphasizes developing accurate and computationally feasible methods, employing architectures like normalizing flows, diffusion models, and Markov Chain Monte Carlo (MCMC) variants, often within a variational inference framework. These advancements are improving uncertainty quantification in machine learning models, enabling more reliable predictions in applications ranging from image processing and inverse problems to scientific modeling and medical image analysis. The ultimate goal is to bridge the gap between the theoretical elegance of Bayesian methods and their practical applicability to high-dimensional, real-world problems.

Papers

May 23, 2023

Amortized Variational Inference with Coverage Guarantees
Yash Patel, Declan McNamara, Jackson Loper, Jeffrey Regier, Ambuj Tewari
Variational Inference Posterior Approximation Amortized Variational Inference

May 15, 2023

Refining Amortized Posterior Approximations using Gradient-Based Summary Statistics
Rafael Orozco, Ali Siahkoohi, Mathias Louboutin, Felix J. Herrmann
Posterior Approximation Bayesian Inverse Problem Amortized Variational Inference Amortized Bayesian Inference

April 21, 2023

April 11, 2023

Black Box Variational Inference with a Deterministic Objective: Faster, More Accurate, and Even More Black Box
Ryan Giordano, Martin Ingram, Tamara Broderick
Variational Inference Black Box Deterministic Algorithm Posterior Approximation Variational Approximation Black Box Variational Inference Mean Field Variational Posterior Covariance

February 8, 2023

Learning How to Infer Partial MDPs for In-Context Adaptation and Exploration
Chentian Jiang, Nan Rosemary Ke, Hado van Hasselt
LeArning Abstract Markov Decision Process Environment Exploration Posterior Sampling Posterior Approximation Exploration Exploitation Contextual Adapter

January 30, 2023

Variational sparse inverse Cholesky approximation for latent Gaussian processes via double Kullback-Leibler minimization
Jian Cao, Myeongjong Kang, Felix Jimenez, Huiyan Sang, Florian Schafer, Matthias Katzfuss
Gaussian Process Posterior Approximation Variational Approximation Sparse Variational Kullback Leibler Cholesky Factorization

January 9, 2023

Constraining cosmological parameters from N-body simulations with Variational Bayesian Neural Networks
Héctor J. Hortúa, Luz Ángela García, Leonardo Castañeda C
Normalizing Flow Astronomical Data Posterior Approximation Variational Bayesian Variational Posterior Cosmological Analysis

December 15, 2022

Bayesian posterior approximation with stochastic ensembles
Oleksandr Balabanov, Bernhard Mehlig, Hampus Linander
Variational Inference Deep Ensemble Posterior Inference Posterior Approximation Stochastic Neural Network Bayesian Ensemble

November 22, 2022

$\beta$-Multivariational Autoencoder for Entangled Representation Learning in Video Frames
Fatemeh Nouri, Robert Bergevin
Variational Autoencoder Video Frame Posterior Approximation Entangled Representation

November 11, 2022

Understanding Approximation for Bayesian Inference in Neural Networks
Sebastian Farquhar
Neural Network Bayesian Inference Posterior Approximation Approximate Inference Approximation Theory Approximate Posterior

November 7, 2022

Posterior samples of source galaxies in strong gravitational lenses with score-based priors
Alexandre Adam, Adam Coogan, Nikolay Malkin, Ronan Legin, Laurence Perreault-Levasseur, Yashar Hezaveh, Yoshua Bengio
Posterior Sampling Posterior Approximation Score Based Distant GALAXY Gravitational Lens

October 26, 2022

Efficient Data Mosaicing with Simulation-based Inference
Andrew Gambardella, Youngjun Choi, Doyo Choi, Jinjoon Lee
Simulation Based Inference Posterior Approximation

September 21, 2022

Improved Marginal Unbiased Score Expansion (MUSE) via Implicit Differentiation
Marius Millea
Posterior Approximation Hierarchical Bayesian Inference Problem Implicit Differentiation MUSE Humor Sub Challenge

September 6, 2022

Bayesian Neural Network Inference via Implicit Models and the Posterior Predictive Distribution
Joel Janek Dabrowski, Daniel Edward Pagendam
Variational Inference Bayesian Neural Network Posterior Inference Posterior Approximation Implicit Model Posterior Predictive Approximate Bayesian Inference Bayesian Neural

September 5, 2022

Investigating the Impact of Model Misspecification in Neural Simulation-based Inference
Patrick Cannon, Daniel Ward, Sebastian M. Schmon
Global Impact Simulation Based Inference Posterior Approximation Model Misspecification Approximate Bayesian Inference Neural Density

July 24, 2022

Reliable amortized variational inference with physics-based latent distribution correction
Ali Siahkoohi, Gabrio Rizzuti, Rafael Orozco, Felix J. Herrmann
Variational Inference Physic Informed Posterior Approximation Amortized Variational Inference High Dimensional Inverse Problem Inversion Problem

June 13, 2022

SIXO: Smoothing Inference with Twisted Objectives
Dieterich Lawson, Allan Raventós, Andrew Warrington, Scott Linderman
Scientific Inference Posterior Approximation Model Learning Sequential Monte Carlo Inference Algorithm

June 8, 2022

Benchmarking Bayesian neural networks and evaluation metrics for regression tasks
Brian Staber, Sébastien Da Veiga
Uncertainty Quantification Bayesian Neural Network Evaluation Metric Posterior Inference Regression Task Posterior Approximation Kernel Stein Discrepancy

May 31, 2022

Likelihood-Free Inference with Generative Neural Networks via Scoring Rule Minimization
Lorenzo Pacchiardi, Ritabrata Dutta
Generative Network Posterior Approximation Generative Neural Network Intractable Likelihood Likelihood Free Inference

Posterior Approximation

Papers

Amortized Variational Inference with Coverage Guarantees

Refining Amortized Posterior Approximations using Gradient-Based Summary Statistics

Machine Learning and the Future of Bayesian Computation

Balancing Simulation-based Inference for Conservative Posteriors

Black Box Variational Inference with a Deterministic Objective: Faster, More Accurate, and Even More Black Box

Learning How to Infer Partial MDPs for In-Context Adaptation and Exploration

Variational sparse inverse Cholesky approximation for latent Gaussian processes via double Kullback-Leibler minimization

Constraining cosmological parameters from N-body simulations with Variational Bayesian Neural Networks

Bayesian posterior approximation with stochastic ensembles

$\beta$-Multivariational Autoencoder for Entangled Representation Learning in Video Frames

Understanding Approximation for Bayesian Inference in Neural Networks

Posterior samples of source galaxies in strong gravitational lenses with score-based priors

Efficient Data Mosaicing with Simulation-based Inference

Improved Marginal Unbiased Score Expansion (MUSE) via Implicit Differentiation

Bayesian Neural Network Inference via Implicit Models and the Posterior Predictive Distribution

Investigating the Impact of Model Misspecification in Neural Simulation-based Inference

Reliable amortized variational inference with physics-based latent distribution correction

SIXO: Smoothing Inference with Twisted Objectives

Benchmarking Bayesian neural networks and evaluation metrics for regression tasks

Likelihood-Free Inference with Generative Neural Networks via Scoring Rule Minimization