Gaussian Posterior

Gaussian posteriors represent a probability distribution over possible solutions to a problem, often used in Bayesian inference to quantify uncertainty. Current research focuses on improving the accuracy and efficiency of Gaussian posterior estimation, particularly in high-dimensional spaces, using techniques like energy-based models, particle filters blended with Gaussian approximations, and modified neural network architectures (e.g., employing Leaky ReLU activations). These advancements are crucial for addressing limitations of simple Gaussian assumptions in complex problems, such as those encountered in robotics (SLAM), machine learning (Bayesian neural networks), and inverse problems, leading to more robust and reliable inferences.

Papers

December 12, 2023

Uncertainty Visualization via Low-Dimensional Posterior Projections
Omer Yair, Elias Nehme, Tomer Michaeli
Posterior Inference Ill Posed Inverse Problem Diffusion Posterior Sampling Uncertainty Visualization Gaussian Posterior

March 24, 2023

GAPSLAM: Blending Gaussian Approximation and Particle Filters for Real-Time Non-Gaussian SLAM
Qiangqiang Huang, John J. Leonard
Simultaneous Localization and Mapping Particle Filter Real Time Inference Gaussian Approximation Gaussian Posterior Gaussian SLAM

March 6, 2023

Bayesian inference with finitely wide neural networks
Chi-Ken Lu
Bayesian Inference Wide Neural Network Gaussian Kernel Non Gaussian Deep Gaussian Process Multivariate Dependence Gaussian Posterior

October 6, 2022

Hypernetwork approach to Bayesian MAML
Piotr Borycki, Piotr Kubacki, Marcin Przewięźlikowski, Tomasz Kuśmierczyk, Jacek Tabor, Przemysław Spurek
Bayesian Learning Model Agnostic Meta Learning Hypernetwork Approach Point Wise Weight Gaussian Posterior

July 26, 2022

One Simple Trick to Fix Your Bayesian Neural Network
Piotr Tempczyk, Ksawery Smoczyński, Philip Smolenski-Jensen, Marek Cygan
Bayesian Neural Network Mean Field ReLU Layer Unconventional Rabbit Hat Trick Neural Posterior Estimation Gaussian Posterior

June 15, 2022

Wide Bayesian neural networks have a simple weight posterior: theory and accelerated sampling
Jiri Hron, Roman Novak, Jeffrey Pennington, Jascha Sohl-Dickstein
Theoretical Understanding Bayesian Neural Network Markov Chain Monte Carlo Wide Neural Network Reparameterization Method Neural Network Gaussian Process Gaussian Posterior

June 8, 2022

Neural Diffusion Processes
Vincent Dutordoir, Alan Saul, Zoubin Ghahramani, Fergus Simpson
Generative Model Neural Process Neural Diffusion Gaussian Posterior

November 15, 2021

ELBD: Efficient score algorithm for feature selection on latent variables of VAE
Yiran Dong, Chuanhou Gao
Feature Selection Latent Variable Lung VAE Score Based Weak Convergence Evidence Lower Bound Gaussian Posterior