The paper extends calibration to sets of probabilistic classifiers, finding many ensembles are poorly calibrated.
problem Evaluating the validity of epistemic uncertainty in sets of probabilistic classifiers.
method Proposed a novel nonparametric calibration test for sets of probabilistic classifiers.
result Ensembles of deep neural networks are often not well calibrated.
A new method for measuring prediction uncertainty in classifiers.
problem Measuring uncertainty of predictions from machine learning methods.
method Density Based Calibration (DBCal) technique.
result Expected calibration error of less than 0.2% on binary classifiers and less than 3% on semantic segmentation networks.
Unified Uncertainty Calibration improves AI predictions by combining different types of uncertainty.
problem AI classifiers struggle with uncertainty, leading to miscalibrated predictions and poor performance.
method Unified Uncertainty Calibration (U2C) combines aleatoric and epistemic uncertainties to improve prediction quality.
result U2C outperforms traditional reject-or-classify methods across various ImageNet benchmarks.
Study on the structure of classifier boundaries in DNA sequencing.
problem Understanding the structure of boundaries in a Bayes classifier for DNA sequencing.
method Examined the structure of the boundary in a Bayes classifier applied to DNA sequencing data. Introduced a new measure of uncertainty, Neighbor Similarity.
result The boundary is large and complex, and Neighbor Similarity effectively measures classifier uncertainty.
New method quantifies classifier uncertainty, revealing large variability in performance metrics.
problem Uncertainty in classifier performance metrics due to small data sets.
method Probability model of the confusion matrix to quantify uncertainty.
result Large uncertainties in classification performance metrics can lead to misleading conclusions.
Paper develops robust SVM classifiers for uncertain data.
problem Sensitivity of SVM classifiers to data uncertainty.
method Two probabilistic approaches: Single Perturbation and Extreme Empirical Loss.
result Both methods reduce data uncertainty effects efficiently.
Rule-based classifiers quantify uncertainty using Bernoulli random variables.
problem Quantifying the uncertainty of precision estimates for rule-based text classifiers.
method Treat partitions of sub-strings as Bernoulli random variables, compare means using statistical tests, and combine classifiers using Dempster-Shafer theory.
result The approach can be used to combine binary classifiers into a multi-label classifier.
ECP method improves image classifier uncertainty sets.
problem Generating reliable uncertainty sets for deep classifiers.
method Evidential Conformal Prediction (ECP) based on EDL.
result ECP outperforms state-of-the-art methods in set size and adaptivity.
Efficient classifier with uncertainty bounds for safety-critical applications.
problem Lack of uncertainty bounds in high-accuracy classifiers for safety-critical tasks.
method Nadaraya-Watson estimator with frequentist bounds.
result Competitive accuracy and uncertainty bounds at reduced computational cost.
Localized uncertainty attacks target uncertain regions to create imperceptible adversarial examples.
problem Adversarial examples that are imperceptible to humans and strong under deterministic classifiers.
method Localized uncertainty attacks by perturbing uncertain regions, using predictive uncertainty or surrogate models.
result Localized uncertainty attacks produce strong adversarial examples that retain input similarity.
Bayesian model fuses multiple classifiers with explicit correlation modeling.
problem Combining outputs of multiple classifiers with explicit correlation.
method Hierarchical Bayesian model with correlated Dirichlet distribution.
result Fused classifier performance can be Bayes optimal even for highly correlated base classifiers.
New method provides formal uncertainty guarantees for image classifiers.
problem Uncertainty quantification for image classifiers without formal guarantees.
method Adapts conformal prediction to give stable, formal coverage guarantees.
result Method outperforms existing approaches in coverage and set size.
We define a generalized likelihood function based on uncertainty measures and show that maximizing such a likelihood function for different measures induces different types of classifiers. In the probabilistic framework, we obtain classifiers that optimize the cross-entropy function. In the possibilistic framework, we …
Fusion of robustness and uncertainty techniques improves adversarial defense.
problem Adversarial attacks on deep neural networks.
method Integrating uncertainty quantification into randomized smoothing for robustness guarantees.
result Improved robustness guarantees for uncertainty aware classifiers.
Characterizes uncertainty in high-dimensional linear classification models.
problem Assessing uncertainty in high-dimensional linear classification models.
method Approximate message passing algorithm for posterior marginals, closed-form formula for joint statistics.
result Closed-form formula for joint statistics between logistic classifier, Bayesian uncertainty, and ground-truth probit uncertainty.
Assistive robots can potentially improve the quality of life and personal independence of elderly people by supporting everyday life activities. To guarantee a safe and intuitive interaction between human and robot, human intentions need to be recognized automatically. As humans communicate their intentions multimodall…
This paper visualizes uncertainty in classifier performance metrics.
problem Overemphasis on model performance metrics risks overlooking uncertainty.
method Developed visualizations of confusion matrix metric distributions.
result Uncertainty in performance metrics can overshadow model differences.
ProBoost boosts probabilistic classifiers by focusing on uncertain samples.
problem Improving probabilistic classifiers through targeted learning.
method ProBoost uses epistemic uncertainty to select challenging samples, increasing their weight for subsequent learners.
result ProBoost significantly improves classifier performance, especially with few weak learners.
We present an analysis of predictive uncertainty based out-of-distribution detection for different approaches to estimate various models' epistemic uncertainty and contrast it with extreme value theory based open set recognition. While the former alone does not seem to be enough to overcome this challenge, we demonstra…
Proposes a method to quantify uncertainty in deterministic image classifiers.
problem Uncertainty in deterministic image classifiers.
method Introduces Wellington Posterior for inductive transfer from scenes.
result Validates Wellington Posterior using various methods.
Novel neural network models quantify uncertainty for deep classifiers.
problem Deep networks' overconfidence and ignorance about uncertainty.
method Variational autoencoders and GANs generate out-of-distribution samples.
result Better uncertainty estimates for in- and out-of-distribution samples.
We study the quantification of uncertainty of Convolutional Neural Networks (CNNs) based on gradient metrics. Unlike the classical softmax entropy, such metrics gather information from all layers of the CNN. We show for the EMNIST digits data set that for several such metrics we achieve the same meta classification acc…
JUCAL jointly calibrates aleatoric and epistemic uncertainties in classifier ensembles.
problem Misrepresentation of predictive uncertainty due to unbalanced aleatoric and epistemic uncertainties.
method Joint Uncertainty Calibration (JUCAL) that jointly calibrates two constants to weight and scale uncertainties.
result Significantly outperforms state-of-the-art calibration methods across various text classification tasks.
New algorithm improves graph-based active learning by identifying unexplored regions.
problem Improving graph-based active learning by identifying unexplored regions.
method Poisson Reweighted Laplacian Uncertainty Sampling (PWLL) with a diagonal perturbation.
result PWLL effectively identifies unexplored regions in graph-based data.
Bayesian framework improves deep classifier reliability.
problem Overconfident models under dataset shift.
method Bayesian inference with out-of-distribution data augmentation.
result Reliable uncertainty estimates for deep classifiers.
A new method quantifies deep neural network uncertainty by mixing OVA and AVA classifiers.
problem Uncertainty quantification in deep neural networks, especially for out-of-distribution data.
method Mixing predictions from OVA and AVA classifiers to improve uncertainty quantification.
result Achieves state-of-the-art performance in quantifying out-of-distribution data.
NCP improves deep classifier uncertainty quantification efficiency.
problem Uncertainty quantification for deep classifiers in high-stake applications.
method Neighborhood Conformal Prediction (NCP) algorithm.
result NCP produces smaller prediction sets than traditional CP methods.
New method calibrates uncertainty estimates for image classifiers without labeled data.
problem Uncertainty estimates for modern classifiers are unreliable without labeled calibration data.
method Calibrates uncertainty estimates using unlabeled examples for distribution shifts.
result Proposes a method that provides excellent uncertainty estimates under natural distribution shifts.
New method corrects biased predictions and uncertainty estimates in classification with nuisance parameters.
problem Tackles biased predictions and invalid uncertainty estimates in classification with nuisance parameters.
method Proposes a method that estimates ROC across the entire nuisance parameter space to devise invariant cutoffs.
result Demonstrates effective domain adaptation and valid prediction sets with high power.
While active learning offers potential cost savings, the actual data efficiency---the reduction in amount of labeled data needed to obtain the same error rate---observed in practice is mixed. This paper poses a basic question: when is active learning actually helpful? We provide an answer for logistic regression with t…
Domain adaptation is an important technique to alleviate performance degradation caused by domain shift, e.g., when training and test data come from different domains. Most existing deep adaptation methods focus on reducing domain shift by matching marginal feature distributions through deep transformations on the inpu…
In this study, we present a multi-class graphical Bayesian predictive classifier that incorporates the uncertainty in the model selection into the standard Bayesian formalism. For each class, the dependence structure underlying the observed features is represented by a set of decomposable Gaussian graphical models. Emp…
Improves uncertainty estimation and OOD detection in neural networks.
problem Accurate uncertainty estimation and OOD detection in neural networks.
method Investigates one-vs-all and distance-based logit representations for probabilities.
result One-vs-all formulations improve calibration without additional complexity.
Simplified plug-in loss approximates EDL for reliable uncertainty estimation.
problem Efficient and reliable uncertainty estimation in real-world sensor-based learning systems.
method Approximate Dirichlet expected objectives with plug-in losses evaluated at the Dirichlet mean.
result Plug-in losses provide comparable predictive accuracy and selective prediction performance to classical EDL, while being simpler to implement.
Many applications of classification methods not only require high accuracy but also reliable estimation of predictive uncertainty. However, while many current classification frameworks, in particular deep neural networks, achieve high accuracy, they tend to incorrectly estimate uncertainty. In this paper, we propose a …
Improves reliability of medical diagnosis uncertainty estimates.
problem Label uncertainty in medical diagnosis.
method Post-hoc alpha-calibration method for neural network classifiers. result Significantly enhances reliability of uncertainty estimates.
This study connects prevalence and machine learning for diagnostic testing.
problem Uncertainty quantification in machine learning for diagnostic tests.
method Developed a numerical homotopy algorithm to estimate classification boundaries and quantify uncertainty.
result The proposed method stabilizes uncertainty quantification in machine learning for diagnostic tests.
A new operator based on t-distributions improves NN classifiers' robustness to out-of-distribution samples.
problem NN classifiers assign extreme probabilities to out-of-distribution samples, leading to unreliable predictions.
method Derive a novel operator using t-distributions to model uncertainty more accurately.
result Classifiers using the new operator are more robust to out-of-distribution samples.
New method compares classifiers using GSD-front, addressing statistical uncertainty and robustness.
problem Comparing classifiers with multiple quality metrics and statistical uncertainty.
method Proposes GSD-front and statistical tests for robust comparisons.
result Reliable method for comparing classifiers with statistical uncertainty and robustness.
Probabilistic classifiers output a probability distribution on target classes rather than just a class prediction. Besides providing a clear separation of prediction and decision making, the main advantage of probabilistic models is their ability to represent uncertainty about predictions. In safety-critical applicatio…
Bayesian autoencoders quantify anomaly uncertainty for safer machine learning.
problem Lack of uncertainty quantification in autoencoders for anomaly detection.
method Formulated Bayesian autoencoders to quantify epistemic and aleatoric anomalies.
result Demonstrated effectiveness of BAEs on benchmark and real datasets.
Paper develops robust k-NN algorithm for few samples.
problem Learning robust classifier from limited samples.
method Distributionally robust formulation of weighted k-NN. result Robust classifier improves generalization with smaller Lipschitz norm.
Proposes HetSNGP method for joint model and data uncertainty modeling.
problem Uncertainty estimation in deep learning for safety-critical applications.
method Jointly models model and data uncertainty with HetSNGP method.
result Outperforms baseline methods on challenging out-of-distribution datasets.
The paper introduces a method to quantify uncertainty in neural networks without parametric assumptions.
problem Uncertainty quantification for neural network predictions.
method Nonparametric estimation of conditional label distribution using Nadaraya-Watson kernel.
result The method effectively disentangles aleatoric and epistemic uncertainties.
Recent advances in machine learning have led to increased deployment of black-box classifiers across a wide variety of applications. In many such situations there is a critical need to both reliably assess the performance of these pre-trained models and to perform this assessment in a label-efficient manner (given that…
Survey on assessing and improving classifier calibration for better decision making.
problem Ensuring classifiers correctly quantify prediction uncertainty.
method Overview of principles, methods, and evaluation metrics for calibration.
result New methods and extensions from binary to multiclass settings.
The issue of disagreements amongst human experts is a ubiquitous one in both machine learning and medicine. In medicine, this often corresponds to doctor disagreements on a patient diagnosis. In this work, we show that machine learning models can be trained to give uncertainty scores to data instances that might result…
OOD detection methods often misidentify OOD points, leading to ineffective safety improvements.
problem Improving model safety through OOD detection methods often leads to incorrect identification of out-of-distribution points.
method Re-examine popular OOD detection procedures based on predictive uncertainty or features of supervised models trained on in-distribution data.
result Popular OOD detection methods incorrectly conflate high uncertainty and far feature-space distance with being out-of-distribution.