Uncertainty Estimation Method

Uncertainty estimation methods aim to quantify the confidence of machine learning models' predictions, crucial for deploying AI in high-stakes applications where reliability is paramount. Current research focuses on improving the accuracy and efficiency of these methods, exploring techniques like deep ensembles, Bayesian neural networks, and evidential deep learning, often tailored to specific model architectures and tasks (e.g., image segmentation, natural language processing). This field is vital for building trustworthy AI systems, addressing concerns about model robustness, out-of-distribution detection, and enabling more reliable decision-making in diverse domains like healthcare and autonomous driving.

Papers

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

November 11, 2022

Comparison of Uncertainty Quantification with Deep Learning in Time Series Regression
Levente Foldesi, Matias Valdenegro-Toro
Deep Learning Human Prediction Time Series Uncertainty Quantification High Uncertainty Anticipation Consistent Comparison Uncertainty Estimation Uncertainty Estimation Method

October 28, 2022

Measuring the Confidence of Traffic Forecasting Models: Techniques, Experimental Comparison and Guidelines towards Their Actionability
Ibai Laña, Ignacio, Olabarrieta, Javier Del Ser
Uncertainty Estimation Barzilai Borwein Technique Traffic Prediction Traffic Forecasting Traffic Data High Confidence Uncertainty Estimation Method Experimental Comparison Actionable Part

October 26, 2022

Uncertainty-based Meta-Reinforcement Learning for Robust Radar Tracking
Julius Ott, Lorenzo Servadei, Gianfranco Mauro, Thomas Stadelmayer, Avik Santra, Robert Wille
Uncertainty Estimation Method Meta RL Robust Radar

October 18, 2022

Uncertainty estimation for out-of-distribution detection in computational histopathology
Lea Goetz
Distribution Detection Uncertainty Estimation Predictive Uncertainty Uncertainty Estimation Method Parameter Uncertainty

October 4, 2022

Uncertainty-Aware Lidar Place Recognition in Novel Environments
Keita Mason, Joshua Knights, Milad Ramezani, Peyman Moghadam, Dimity Miller
Uncertainty Estimation Place Recognition Uncertainty Estimation Method New Environment

October 2, 2022

Uncertainty estimations methods for a deep learning model to aid in clinical decision-making -- a clinician's perspective
Michael Dohopolski, Kai Wang, Biling Wang, Ti Bai, Dan Nguyen, David Sher, Steve Jiang, Jing Wang
Deep Learning Model Visual Perspective Uncertainty Estimation Method Uncertainty Prediction Clinical Decision Estimation Method Reliable Prediction Deep Uncertainty

July 19, 2022

Uncertainty in Contrastive Learning: On the Predictability of Downstream Performance
Shervin Ardeshir, Navid Azizan
Contrastive Learning High Uncertainty Anticipation State of the Art Deep Predictability Problem Uncertainty Estimation Method Contrastive Pretraining Downstream Performance

June 22, 2022

Influence of uncertainty estimation techniques on false-positive reduction in liver lesion detection
Ishaan Bhat, Josien P. W. Pluim, Max A. Viergever, Hugo J. Kuijf
Neural Network External Influence False Positive Uncertainty Estimation Method LIver Lesion False Negative Prediction

April 29, 2022

Tailored Uncertainty Estimation for Deep Learning Systems
Joachim Sicking, Maram Akila, Jan David Schneider, Fabian Hüger, Peter Schlicht, Tim Wirtz, Stefan Wrobel
Uncertainty Quantification High Uncertainty Anticipation Uncertainty Estimation Deep Learning System Uncertainty Estimation Method Estimation Performance

April 25, 2022

Data Uncertainty without Prediction Models
Bongjoon Park, Eunkyung Koh
Machine Learning Uncertainty Estimation Prediction Model Uncertainty Estimation Method Data Uncertainty

March 16, 2022

Layer Ensembles: A Single-Pass Uncertainty Estimation in Deep Learning for Segmentation
Kaisar Kushibar, Víctor Manuel Campello, Lidia Garrucho Moras, Akis Linardos, Petia Radeva, Karim Lekadir
Deep Learning Uncertainty Quantification Segmentation Based Approach Uncertainty Estimation Deep Ensemble Pixel Wise Uncertainty Uncertainty Estimation Method Layer Ensemble

February 3, 2022

Maximum Likelihood Uncertainty Estimation: Robustness to Outliers
Deebul S. Nair, Nico Hochgeschwender, Miguel A. Olivares-Mendez
Native Robustness Uncertainty Estimation Distribution Data Global Outlier Uncertainty Estimation Method Better Uncertainty Estimation

November 18, 2021

Exploring the Limits of Epistemic Uncertainty Quantification in Low-Shot Settings
Matias Valdenegro-Toro
Uncertainty Quantification Distribution Detection AI System Continuum Limit Epistemic Uncertainty Uncertainty Estimation Method Epistemic Uncertainty Quantification

Uncertainty Estimation Method

Papers

Looking at the posterior: accuracy and uncertainty of neural-network predictions

Comparison of Uncertainty Quantification with Deep Learning in Time Series Regression

Measuring the Confidence of Traffic Forecasting Models: Techniques, Experimental Comparison and Guidelines towards Their Actionability

Uncertainty-based Meta-Reinforcement Learning for Robust Radar Tracking

Uncertainty estimation for out-of-distribution detection in computational histopathology

Uncertainty-Aware Lidar Place Recognition in Novel Environments

Uncertainty estimations methods for a deep learning model to aid in clinical decision-making -- a clinician's perspective

Uncertainty in Contrastive Learning: On the Predictability of Downstream Performance

Influence of uncertainty estimation techniques on false-positive reduction in liver lesion detection

Tailored Uncertainty Estimation for Deep Learning Systems

Data Uncertainty without Prediction Models

Layer Ensembles: A Single-Pass Uncertainty Estimation in Deep Learning for Segmentation

Maximum Likelihood Uncertainty Estimation: Robustness to Outliers

Exploring the Limits of Epistemic Uncertainty Quantification in Low-Shot Settings