Posterior Network

Posterior networks are neural network architectures designed to learn and represent probability distributions, particularly posterior distributions in Bayesian inference. Current research focuses on improving their accuracy and efficiency in various applications, including uncertainty quantification in graph neural networks and image reconstruction, often employing techniques like normalizing flows, energy-based models, and diffusion processes to enhance model expressivity and sampling. These advancements are significant for addressing challenges in high-dimensional data analysis and improving the reliability of predictions in diverse fields such as medical imaging and federated learning, where robust uncertainty estimates are crucial.

Papers

October 24, 2024

Enhancing Feature-Specific Data Protection via Bayesian Coordinate Differential Privacy
Maryam Aliakbarpour, Syomantak Chaudhuri, Thomas A. Courtade, Alireza Fallah, Michael I. Jordan
Differential Privacy Privacy Guarantee Local Differential Privacy Posterior Network Private Bayesian

October 21, 2024

BI-EqNO: Generalized Approximate Bayesian Inference with an Equivariant Neural Operator Framework
Xu-Hui Zhou, Zhuo-Ran Liu, Heng Xiao
Neural Operator Bayesian Inference Data Assimilation Equivariant Map Approximate Bayesian Inference Ensemble Kalman Filter Posterior Network

September 6, 2024

CUQ-GNN: Committee-based Graph Uncertainty Quantification using Posterior Networks
Clemens Damke, Eyke Hüllermeier
Graph Neural Network Predictive Uncertainty Uncertainty Aware Graph Posterior Network

May 22, 2024

Learning Latent Space Hierarchical EBM Diffusion Models
Jiali Cui, Tian Han
Latent Diffusion Prior Model Posterior Network Energy Based Prior

February 8, 2024

Memory-efficient deep end-to-end posterior network (DEEPEN) for inverse problems
Jyothi Rikhab Chand, Mathews Jacob
Inverse Problem Posterior Inference Image Recovery Magnetic Resonance Image Reconstruction Posterior Network Memory effIcieNt Unrolled Optimization

December 18, 2023

Dirichlet-based Uncertainty Quantification for Personalized Federated Learning with Improved Posterior Networks
Nikita Kotelevskii, Samuel Horváth, Karthik Nandakumar, Martin Takáč, Maxim Panov
Federated Learning Real World Personalized Federated Learning Predictive Uncertainty Global Model Personalized Model Posterior Network

November 10, 2023

Improvements on Uncertainty Quantification for Node Classification via Distance-Based Regularization
Russell Alan Hart, Linlin Yu, Yifei Lou, Feng Chen
Uncertainty Quantification Node Classification Cross Entropy Loss Large Relevance Improvement Posterior Network Distance Based Regularization

June 8, 2023

Unscented Autoencoder
Faris Janjoš, Lars Rosenbaum, Maxim Dolgov, J. Marius Zöllner
Variational Autoencoder Unscented Kalman Filter Adversarial AutoEncoder Posterior Network Latent Posterior

February 17, 2023

JANA: Jointly Amortized Neural Approximation of Complex Bayesian Models
Stefan T. Radev, Marvin Schmitt, Valentin Pratz, Umberto Picchini, Ullrich Köthe, Paul-Christian Bürkner
Amortized Bayesian Inference Posterior Network Neural Likelihood Complex Posterior

October 6, 2022

Residual-based error correction for neural operator accelerated infinite-dimensional Bayesian inverse problems
Lianghao Cao, Thomas O'Leary-Roseberry, Prashant K. Jha, J. Tinsley Oden, Omar Ghattas
Inverse Problem Neural Operator Error Correction Posterior Network

May 30, 2022

Deep Posterior Distribution-based Embedding for Hyperspectral Image Super-resolution
Jinhui Hou, Zhiyu Zhu, Junhui Hou, Huanqiang Zeng, Jinjian Wu, Jiantao Zhou
Super Resolution Hyperspectral Image Hyperspectral Image Super Resolution Spatial Spectral Information Posterior Network

February 9, 2022

Covariate-informed Representation Learning to Prevent Posterior Collapse of iVAE
Young-geun Kim, Ying Liu, Xuexin Wei
Latent Representation Variational AutoEncoder Posterior Collapse Posterior Network Covariate Representation

January 18, 2022

Online, Informative MCMC Thinning with Kernelized Stein Discrepancy
Cole Hawkins, Alec Koppel, Zheng Zhang
Bayesian Inference Markov Chain Monte Carlo Online Service Online Bayesian Informative Prior Posterior Network Stein Discrepancy

Posterior Network

Papers

Enhancing Feature-Specific Data Protection via Bayesian Coordinate Differential Privacy

BI-EqNO: Generalized Approximate Bayesian Inference with an Equivariant Neural Operator Framework

CUQ-GNN: Committee-based Graph Uncertainty Quantification using Posterior Networks

Learning Latent Space Hierarchical EBM Diffusion Models

Memory-efficient deep end-to-end posterior network (DEEPEN) for inverse problems

Dirichlet-based Uncertainty Quantification for Personalized Federated Learning with Improved Posterior Networks

Improvements on Uncertainty Quantification for Node Classification via Distance-Based Regularization

Unscented Autoencoder

JANA: Jointly Amortized Neural Approximation of Complex Bayesian Models

Residual-based error correction for neural operator accelerated infinite-dimensional Bayesian inverse problems

Deep Posterior Distribution-based Embedding for Hyperspectral Image Super-resolution

Covariate-informed Representation Learning to Prevent Posterior Collapse of iVAE

Online, Informative MCMC Thinning with Kernelized Stein Discrepancy