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

May 25, 2023

Bi-fidelity Variational Auto-encoder for Uncertainty Quantification
Nuojin Cheng, Osman Asif Malik, Subhayan De, Stephen Becker, Alireza Doostan
Uncertainty Quantification Information Bottleneck Variational Auto VAE Model

May 24, 2023

Timeseries-aware Uncertainty Wrappers for Uncertainty Quantification of Information-Fusion-Enhanced AI Models based on Machine Learning
Janek Groß, Michael Kläs, Lisa Jöckel, Pascal Gerber
Machine Learning Uncertainty Quantification Artificial Intelligence Model Information Fusion Input Feature Reliable Uncertainty Uncertainty Wrapper

May 23, 2023

May 22, 2023

Cycle Consistency-based Uncertainty Quantification of Neural Networks in Inverse Imaging Problems
Luzhe Huang, Jianing Li, Xiaofu Ding, Yijie Zhang, Hanlong Chen, Aydogan Ozcan
Neural Network Uncertainty Quantification Inverse Problem Uncertainty Estimation

May 20, 2023

Distribution-Free Model-Agnostic Regression Calibration via Nonparametric Methods
Shang Liu, Zhongze Cai, Xiaocheng Li
Uncertainty Quantification Nonparametric Approach Individual Calibration

May 18, 2023

Uncertainty Quantification in Deep Neural Networks through Statistical Inference on Latent Space
Luigi Sbailò, Luca M. Ghiringhelli
Deep Neural Network Latent Space Uncertainty Quantification Distribution Data Statistical Inference Latent Space Representation Deep Learning Classifier Overconfidence Problem

May 17, 2023

Principal Uncertainty Quantification with Spatial Correlation for Image Restoration Problems
Omer Belhasin, Yaniv Romano, Daniel Freedman, Ehud Rivlin, Michael Elad
Uncertainty Quantification Spatial Correlation Image Restoration Problem Uncertainty Region

May 15, 2023

Learning Switching Port-Hamiltonian Systems with Uncertainty Quantification
Thomas Beckers, Tom Z. Jiahao, George J. Pappas
LeArning Abstract Gaussian Process Uncertainty Quantification Physical System Port Hamiltonian

May 10, 2023

ST-GIN: An Uncertainty Quantification Approach in Traffic Data Imputation with Spatio-temporal Graph Attention and Bidirectional Recurrent United Neural Networks
Zepu Wang, Dingyi Zhuang, Yankai Li, Jinhua Zhao, Peng Sun, Shenhao Wang, Yulin Hu
Neural Network Deep Learning Uncertainty Quantification Graph Attention Traffic Data Bidirectional Recurrent Neural Network Traffic Data Imputation

May 7, 2023

Uncertainty Quantification in Machine Learning for Engineering Design and Health Prognostics: A Tutorial
Venkat Nemani, Luca Biggio, Xun Huan, Zhen Hu, Olga Fink, Anh Tran, Yan Wang, Xiaoge Zhang, Chao Hu
Machine Learning Uncertainty Quantification Predictive Uncertainty Tutorial Review Prognostic Prediction Useful Life Engineering Design Data Driven Prognostic

May 4, 2023

Self-Supervised Learning for Organs At Risk and Tumor Segmentation with Uncertainty Quantification
Ilkin Isler, Debesh Jha, Curtis Lisle, Justin Rineer, Patrick Kelly, Bulent Aydogan, Mohamed Abazeed, Damla Turgut, Ulas Bagci
Self Supervised Learning Uncertainty Quantification Tumor Segmentation Self Supervised Pre Training High Quality Risk Description Human Organ

May 2, 2023

Single-model uncertainty quantification in neural network potentials does not consistently outperform model ensembles
Aik Rui Tan, Shingo Urata, Samuel Goldman, Johannes C. B. Dietschreit, Rafael Gómez-Bombarelli
Uncertainty Quantification Interatomic Potential Atomistic Simulation Model Ensemble Neural Network Potential Based NIDS

May 1, 2023

April 27, 2023

April 25, 2023

Latent diffusion models for generative precipitation nowcasting with accurate uncertainty quantification
Jussi Leinonen, Ulrich Hamann, Daniele Nerini, Urs Germann, Gabriele Franch
Diffusion Model Uncertainty Quantification Latent Diffusion Model Generative Neural Network Precipitation Nowcasting Precipitation Forecasting

April 23, 2023

Quantile Extreme Gradient Boosting for Uncertainty Quantification
Xiaozhe Yin, Masoud Fallah-Shorshani, Rob McConnell, Scott Fruin, Yao-Yi Chiang, Meredith Franklin
Uncertainty Quantification Quantile Regression Extreme Gradient Boosting

April 17, 2023

Goal-oriented Uncertainty Quantification for Inverse Problems via Variational Encoder-Decoder Networks
Babak Maboudi Afkham, Julianne Chung, Matthias Chung
Uncertainty Quantification Inverse Problem Variational Encoder Nonlinear CT Image Reconstruction

Uncertainty Quantification

Papers

Bi-fidelity Variational Auto-encoder for Uncertainty Quantification

Timeseries-aware Uncertainty Wrappers for Uncertainty Quantification of Information-Fusion-Enhanced AI Models based on Machine Learning

Uncertainty Quantification over Graph with Conformalized Graph Neural Networks

UPNet: Uncertainty-based Picking Deep Learning Network for Robust First Break Picking

Cycle Consistency-based Uncertainty Quantification of Neural Networks in Inverse Imaging Problems

Distribution-Free Model-Agnostic Regression Calibration via Nonparametric Methods

Uncertainty Quantification in Deep Neural Networks through Statistical Inference on Latent Space

Principal Uncertainty Quantification with Spatial Correlation for Image Restoration Problems

Learning Switching Port-Hamiltonian Systems with Uncertainty Quantification

ST-GIN: An Uncertainty Quantification Approach in Traffic Data Imputation with Spatio-temporal Graph Attention and Bidirectional Recurrent United Neural Networks

Uncertainty Quantification in Machine Learning for Engineering Design and Health Prognostics: A Tutorial

Self-Supervised Learning for Organs At Risk and Tumor Segmentation with Uncertainty Quantification

Single-model uncertainty quantification in neural network potentials does not consistently outperform model ensembles

Probabilistic 3D segmentation for aleatoric uncertainty quantification in full 3D medical data

Proper Scoring Rules for Survival Analysis

Deep Ensembles to Improve Uncertainty Quantification of Statistical Downscaling Models under Climate Change Conditions

Uncertainty Aware Neural Network from Similarity and Sensitivity

Latent diffusion models for generative precipitation nowcasting with accurate uncertainty quantification

Quantile Extreme Gradient Boosting for Uncertainty Quantification

Goal-oriented Uncertainty Quantification for Inverse Problems via Variational Encoder-Decoder Networks