Accurate Uncertainty

Accurate uncertainty quantification in machine learning aims to provide reliable estimates of prediction confidence, crucial for trustworthy decision-making in high-stakes applications. Current research focuses on developing and improving methods for calibrated uncertainty estimation across diverse model architectures, including neural networks (e.g., U-Nets, Bayesian neural networks), Gaussian processes, and ensemble methods, often employing techniques like conformal prediction and recalibration strategies to address issues like distribution shifts and overconfidence. This work is significant because accurate uncertainty estimates enhance the reliability and safety of machine learning systems in various fields, from medical diagnosis and autonomous driving to energy systems and scientific simulations.

Papers

September 27, 2024

Calibrated Probabilistic Forecasts for Arbitrary Sequences
Charles Marx, Volodymyr Kuleshov, Stefano Ermon
Probabilistic Forecast Predictive Performance Random Sequence Accurate Uncertainty Approachability Problem

August 23, 2024

SUMO: Search-Based Uncertainty Estimation for Model-Based Offline Reinforcement Learning
Zhongjian Qiao, Jiafei Lyu, Kechen Jiao, Qi Liu, Xiu Li
Offline Reinforcement Learning Robust Reinforcement Learning Model Based Offline Reinforcement Learning Accurate Uncertainty

July 15, 2024

Expert-aware uncertainty estimation for quality control of neural-based blood typing
Ekaterina Zaychenkova, Dmitrii Iarchuk, Sergey Korchagin, Alexey Zaitsev, Egor Ershov
Ground Truth Uncertainty Estimation Uncertainty Prediction Quality Control Accurate Uncertainty Consensus Value

June 6, 2024

Energy-based Epistemic Uncertainty for Graph Neural Networks
Dominik Fuchsgruber, Tom Wollschläger, Stephan Günnemann
Graph Neural Network Epistemic Uncertainty Energy Based Model Accurate Uncertainty Structural Uncertainty

May 3, 2024

A comparative study of conformal prediction methods for valid uncertainty quantification in machine learning
Nicolas Dewolf
Machine Learning Comparative Study Uncertainty Quantification Conformal Prediction Predictive Model Bayesian Perspective Accurate Uncertainty

February 26, 2024

Pretrained Visual Uncertainties
Michael Kirchhof, Mark Collier, Seong Joon Oh, Enkelejda Kasneci
Large Scale Pretraining Accurate Uncertainty Uncertainty Visualization

December 16, 2023

Identity Curvature Laplace Approximation for Improved Out-of-Distribution Detection
Maksim Zhdanov, Stanislav Dereka, Sergey Kolesnikov
Uncertainty Estimation Bayesian Approach Understanding Out of Distribution Accurate Uncertainty Laplace Approximation

December 4, 2023

Instant Uncertainty Calibration of NeRFs Using a Meta-Calibrator
Niki Amini-Naieni, Tomas Jakab, Andrea Vedaldi, Ronald Clark
Neural Radiance Field Uncertainty Calibration NeRF Method Uncertainty Map Accurate Uncertainty

October 25, 2023

Adaptive Uncertainty Estimation via High-Dimensional Testing on Latent Representations
Tsai Hor Chan, Kin Wai Lau, Jiajun Shen, Guosheng Yin, Lequan Yu
High Dimensional Latent Representation Uncertainty Estimation Uncertainty Estimation Method Accurate Uncertainty Uncertainty Estimation Performance Adaptive Uncertainty

August 21, 2023

Deep Evidential Learning for Bayesian Quantile Regression
Frederik Boe Hüttel, Filipe Rodrigues, Francisco Câmara Pereira
Uncertainty Quantification Epistemic Uncertainty Quantile Regression Non Gaussian Accurate Uncertainty

August 14, 2023

Probabilistic MIMO U-Net: Efficient and Accurate Uncertainty Estimation for Pixel-wise Regression
Anton Baumann, Thomas Roßberg, Michael Schmitt
Deep Neural Network Uncertainty Estimation U Net Accurate Uncertainty Pixel Wise Regression

August 6, 2023

Building Safe and Reliable AI systems for Safety Critical Tasks with Vision-Language Processing
Shuang Ao
Artificial Intelligence Uncertainty Quantification Model Uncertainty Accurate Uncertainty System Safety Safety Critical Task

July 5, 2023

Distance Preserving Machine Learning for Uncertainty Aware Accelerator Capacitance Predictions
Steven Goldenberg, Malachi Schram, Kishansingh Rajput, Thomas Britton, Chris Pappas, Dan Lu, Jared Walden, Majdi I. Radaideh, Sarah Cousineau, Sudarshan Harave
Gaussian Process Phantom 2D Accelerator Accurate Uncertainty Capacitive Sensing Distance Preserving

February 6, 2023

Surrogate uncertainty estimation for your time series forecasting black-box: learn when to trust
Leonid Erlygin, Vladimir Zholobov, Valeriia Baklanova, Evgeny Sokolovskiy, Alexey Zaytsev
Gaussian Process Time Series Forecasting Uncertainty Estimation Surrogate Model Robust Model Accurate Uncertainty

January 11, 2023

Uncertainty Estimation based on Geometric Separation
Gabriella Chouraqui, Liron Cohen, Gil Einziger, Liel Leman
High Uncertainty Anticipation Uncertainty Estimation Geometric Approach Accurate Uncertainty Point to Plane Distance Spatial Separation

October 9, 2022

Test-time Recalibration of Conformal Predictors Under Distribution Shift Based on Unlabeled Examples
Fatih Furkan Yilmaz, Reinhard Heckel
Distribution Shift Uncertainty Estimation Image Classifier Accurate Uncertainty Conformal Predictor Unlabeled Example Calibration Set

June 15, 2022

CRISP - Reliable Uncertainty Estimation for Medical Image Segmentation
Thierry Judge, Olivier Bernard, Mihaela Porumb, Agis Chartsias, Arian Beqiri, Pierre-Marc Jodoin
Medical Image Segmentation Pixel Wise Uncertainty Reliable Uncertainty Uncertainty Map Accurate Uncertainty Joint Latent Space

May 15, 2022

Evaluating Uncertainty Calibration for Open-Set Recognition
Zongyao Lyu, Nolan B. Gutierrez, William J. Beksi
Open Set Recognition Uncertainty Calibration Accurate Uncertainty Neural Network Calibration

May 4, 2022

Multivariate Prediction Intervals for Random Forests
Brendan Folie, Maxwell Hutchinson
Random Forest Sequential Learning Prediction Interval Accurate Uncertainty Iterative Design

December 8, 2021

Online Calibrated and Conformal Prediction Improves Bayesian Optimization
Shachi Deshpande, Charles Marx, Volodymyr Kuleshov
Bayesian Optimization Conformal Prediction Sequential Decision Making Task Online Calibration Accurate Uncertainty

Accurate Uncertainty

Papers

Calibrated Probabilistic Forecasts for Arbitrary Sequences

SUMO: Search-Based Uncertainty Estimation for Model-Based Offline Reinforcement Learning

Expert-aware uncertainty estimation for quality control of neural-based blood typing

Energy-based Epistemic Uncertainty for Graph Neural Networks

A comparative study of conformal prediction methods for valid uncertainty quantification in machine learning

Pretrained Visual Uncertainties

Identity Curvature Laplace Approximation for Improved Out-of-Distribution Detection

Instant Uncertainty Calibration of NeRFs Using a Meta-Calibrator

Adaptive Uncertainty Estimation via High-Dimensional Testing on Latent Representations

Deep Evidential Learning for Bayesian Quantile Regression

Probabilistic MIMO U-Net: Efficient and Accurate Uncertainty Estimation for Pixel-wise Regression

Building Safe and Reliable AI systems for Safety Critical Tasks with Vision-Language Processing

Distance Preserving Machine Learning for Uncertainty Aware Accelerator Capacitance Predictions

Surrogate uncertainty estimation for your time series forecasting black-box: learn when to trust

Uncertainty Estimation based on Geometric Separation

Test-time Recalibration of Conformal Predictors Under Distribution Shift Based on Unlabeled Examples

CRISP - Reliable Uncertainty Estimation for Medical Image Segmentation

Evaluating Uncertainty Calibration for Open-Set Recognition

Multivariate Prediction Intervals for Random Forests

Online Calibrated and Conformal Prediction Improves Bayesian Optimization