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

July 2, 2023

Morse Neural Networks for Uncertainty Quantification
Benoit Dherin, Huiyi Hu, Jie Ren, Michael W. Dusenberry, Balaji Lakshminarayanan
Generative Model Uncertainty Quantification Generative Modeling Unnormalized Density Morse Theory

June 27, 2023

Evaluation of machine learning architectures on the quantification of epistemic and aleatoric uncertainties in complex dynamical systems
Stephen Guth, Alireza Mojahed, Themistoklis P. Sapsis
Global Evaluation Uncertainty Quantification Surrogate Model Bayesian Neural Network Aleatoric Uncertainty Complex Dynamical System Model Accuracy Machine Learning Architecture Data Driven Reduced Order

June 26, 2023

Enhanced multi-fidelity modelling for digital twin and uncertainty quantification
AS Desai, Navaneeth N, S Adhikari, S Chakraborty
Uncertainty Quantification Digital Twin Multi Fidelity Fidelity Model Multi Fidelity Surrogate

June 16, 2023

Uncertainty Quantification via Spatial-Temporal Tweedie Model for Zero-inflated and Long-tail Travel Demand Prediction
Xinke Jiang, Dingyi Zhuang, Xianghui Zhang, Hao Chen, Jiayuan Luo, Xiaowei Gao
Uncertainty Quantification Spatio Temporal Temporal Graph Neural Network Travel Demand Travel Time Zero Inflated Sale Pattern

June 13, 2023

User-defined Event Sampling and Uncertainty Quantification in Diffusion Models for Physical Dynamical Systems
Marc Finzi, Anudhyan Boral, Andrew Gordon Wilson, Fei Sha, Leonardo Zepeda-Núñez
Diffusion Model Uncertainty Quantification Stochastic Approximation Physical System Probabilistic Generative Model Conditional Sampling

June 9, 2023

Efficient Uncertainty Quantification and Reduction for Over-Parameterized Neural Networks
Ziyi Huang, Henry Lam, Haofeng Zhang
Neural Network Uncertainty Quantification Neural Tangent Kernel Deep Ensemble Constructive Reduction High Uncertainty Value

June 5, 2023

Uncertainty in Natural Language Processing: Sources, Quantification, and Applications
Mengting Hu, Zhen Zhang, Shiwan Zhao, Minlie Huang, Bingzhe Wu
Natural Language Processing Financial Application Uncertainty Quantification High Uncertainty Anticipation Uncertainty Estimation Source Table

June 1, 2023

May 31, 2023

May 29, 2023

Beyond Confidence: Reliable Models Should Also Consider Atypicality
Mert Yuksekgonul, Linjun Zhang, James Zou, Carlos Guestrin
Uncertainty Quantification High Confidence Low Confidence Ambiguous Input Model Reliability Atypical Speech Rare Class

May 28, 2023

May 27, 2023

Federated Conformal Predictors for Distributed Uncertainty Quantification
Charles Lu, Yaodong Yu, Sai Praneeth Karimireddy, Michael I. Jordan, Ramesh Raskar
Federated Learning Uncertainty Quantification Conformal Prediction

May 26, 2023

Vecchia Gaussian Process Ensembles on Internal Representations of Deep Neural Networks
Felix Jimenez, Matthias Katzfuss
Deep Neural Network Gaussian Process Uncertainty Quantification Internal Representation Robust Prediction

Uncertainty Quantification

Papers

Morse Neural Networks for Uncertainty Quantification

Evaluation of machine learning architectures on the quantification of epistemic and aleatoric uncertainties in complex dynamical systems

Enhanced multi-fidelity modelling for digital twin and uncertainty quantification

Uncertainty Quantification via Spatial-Temporal Tweedie Model for Zero-inflated and Long-tail Travel Demand Prediction

User-defined Event Sampling and Uncertainty Quantification in Diffusion Models for Physical Dynamical Systems

Efficient Uncertainty Quantification and Reduction for Over-Parameterized Neural Networks

Uncertainty in Natural Language Processing: Sources, Quantification, and Applications

A General Framework for Uncertainty Quantification via Neural SDE-RNN

Quantifying Deep Learning Model Uncertainty in Conformal Prediction

Deep Operator Learning-based Surrogate Models with Uncertainty Quantification for Optimizing Internal Cooling Channel Rib Profiles

Probabilistic Uncertainty Quantification of Prediction Models with Application to Visual Localization

DiffLoad: Uncertainty Quantification in Electrical Load Forecasting with the Diffusion Model

Probabilistic computation and uncertainty quantification with emerging covariance

Generating with Confidence: Uncertainty Quantification for Black-box Large Language Models

Perturbation-Assisted Sample Synthesis: A Novel Approach for Uncertainty Quantification

Beyond Confidence: Reliable Models Should Also Consider Atypicality

Conformal Prediction with Large Language Models for Multi-Choice Question Answering

Speech Intelligibility Assessment of Dysarthric Speech by using Goodness of Pronunciation with Uncertainty Quantification

Federated Conformal Predictors for Distributed Uncertainty Quantification

Vecchia Gaussian Process Ensembles on Internal Representations of Deep Neural Networks