Predictive Density

Predictive density estimation aims to generate a probability distribution representing the uncertainty in a model's predictions, going beyond simple point estimates. Current research focuses on improving the efficiency and accuracy of predictive density estimation using various methods, including Gaussian processes, Bayesian deep neural networks, and sophisticated approximation techniques like Gauss-Hermite quadrature to handle high-dimensional or correlated data. These advancements are crucial for reliable uncertainty quantification in diverse scientific fields and applications, enabling more informed decision-making in areas such as multi-objective optimization and scientific modeling where understanding prediction uncertainty is paramount.

Papers

October 8, 2024

Temperature Optimization for Bayesian Deep Learning
Kenyon Ng, Chris van der Heide, Liam Hodgkinson, Susan Wei
Bayesian Deep Learning Thermal Management Posterior Predictive Critical Temperature Cold Posterior Predictive Density

March 3, 2024

Fusion of Gaussian Processes Predictions with Monte Carlo Sampling
Marzieh Ajirak, Daniel Waxman, Fernando Llorente, Petar M. Djuric
Gaussian Process Hybrid Fusion Pooling Layer Predictive Distribution Monte Carlo Sampling Predictive Density

June 15, 2022

Efficient Approximation of Expected Hypervolume Improvement using Gauss-Hermite Quadrature
Alma Rahat, Tinkle Chugh, Jonathan Fieldsend, Richard Allmendinger, Kaisa Miettinen
Probabilistic Surrogate Efficient Approximation Multi Objective Optimisation Hypervolume Approximation Predictive Density

June 12, 2022

Density Regression and Uncertainty Quantification with Bayesian Deep Noise Neural Networks
Daiwei Zhang, Tianci Liu, Jian Kang
Uncertainty Quantification Latent Noise Density Regression Predictive Density

December 13, 2021

How Good are Low-Rank Approximations in Gaussian Process Regression?
Constantinos Daskalakis, Petros Dellaportas, Aristeidis Panos
Gaussian Process Average Approximation Gaussian Process Regression Low Rank Approximation Low Rank Kernel Predictive Density

Predictive Density

Papers

Temperature Optimization for Bayesian Deep Learning

Fusion of Gaussian Processes Predictions with Monte Carlo Sampling

Efficient Approximation of Expected Hypervolume Improvement using Gauss-Hermite Quadrature

Density Regression and Uncertainty Quantification with Bayesian Deep Noise Neural Networks

How Good are Low-Rank Approximations in Gaussian Process Regression?