Uncertainty Quantification

Uncertainty quantification (UQ) aims to assess and represent the confidence in predictions made by machine learning models, crucial for high-stakes applications where reliable predictions are paramount. Current research focuses on developing robust UQ methods, particularly addressing biases in predictions and efficiently quantifying uncertainty in large language models and deep neural networks, often employing techniques like conformal prediction, Bayesian methods, and ensemble learning. The ability to reliably quantify uncertainty enhances the trustworthiness and applicability of machine learning across diverse fields, from healthcare diagnostics and autonomous driving to climate modeling and drug discovery.

Papers

February 7, 2023

IB-UQ: Information bottleneck based uncertainty quantification for neural function regression and neural operator learning
Ling Guo, Hao Wu, Wenwen Zhou, Yan Wang, Tao Zhou
Uncertainty Quantification Neural Operator Information Bottleneck Deep Regression Confidence Map Uncertainty Quantification Method Poor Extrapolation Performance

February 2, 2023

January 18, 2023

Physics-informed Information Field Theory for Modeling Physical Systems with Uncertainty Quantification
Alex Alberts, Ilias Bilionis
Uncertainty Quantification Langevin Dynamic Model Uncertainty Physical System Information Field Theory

January 13, 2023

December 28, 2022

A Learning-Based Optimal Uncertainty Quantification Method and Its Application to Ballistic Impact Problems
Xingsheng Sun, Burigede Liu
Application Proficiency Uncertainty Quantification Parameter Uncertainty Probability Measure Uncertainty Information

December 24, 2022

Improving Uncertainty Quantification of Variance Networks by Tree-Structured Learning
Wenxuan Ma, Xing Yan, Kun Zhang
Neural Network Uncertainty Quantification Tree Learning Variance Network

December 20, 2022

Uncertainty Quantification of MLE for Entity Ranking with Covariates
Jianqing Fan, Jikai Hou, Mengxin Yu
Uncertainty Quantification Maximum Likelihood Latent Feature Relevant Covariates Ranking Performance

December 15, 2022

Scalable Bayesian Uncertainty Quantification for Neural Network Potentials: Promise and Pitfalls
Stephan Thaler, Gregor Doehner, Julija Zavadlav
Uncertainty Quantification Common Pitfall Molecular Dynamic Coarse Grained Promise Neural Network Potential

December 14, 2022

December 13, 2022

Dual Accuracy-Quality-Driven Neural Network for Prediction Interval Generation
Giorgio Morales, John W. Sheppard
Uncertainty Quantification Prediction Interval Target Prediction Regression Network Quality Regression Network

December 8, 2022

Enhanced prediction accuracy with uncertainty quantification in monitoring CO2 sequestration using convolutional neural networks
Yanhua Liu, Xitong Zhang, Ilya Tsvankin, Youzuo Lin
Convolutional Neural Network Uncertainty Quantification Prediction Accuracy Real Data Carbon Capture Seismic Inversion Carbon Sequestration

November 29, 2022

UQ-ARMED: Uncertainty quantification of adversarially-regularized mixed effects deep learning for clustered non-iid data
Alex Treacher, Kevin Nguyen, Dylan Owens, Daniel Heitjan, Albert Montillo
Uncertainty Quantification Ensemble Model Ensemble Method Non IID Data Explainable Metric

November 26, 2022

Looking at the posterior: accuracy and uncertainty of neural-network predictions
H. Linander, O. Balabanov, H. Yang, B. Mehlig
Uncertainty Quantification High Uncertainty Anticipation Bayesian Inference Aleatoric Uncertainty Uncertainty Estimation Method Neural Network Prediction