New method improves model calibration efficiency and accuracy.
problem Improving model calibration for better probability estimates.
method Scaling-binning calibrator method that reduces variance and ensures calibration.
result 35% lower calibration error than histogram binning and guarantees true calibration.
This paper calibrates uncertainty in dropout variational inference models.
problem Uncertainty in variational inference with dropout is poorly calibrated.
method Temperature scaling is extended to dropout variational inference.
result Temperature scaling reduces miscalibration of uncertainty.
LiST improves neural network robustness and calibration without manual tuning.
problem Developing robust and calibrated neural networks simultaneously.
method Lipschitz Scaling Training (LiST) that iteratively adjusts the global Lipschitz constant.
result LiST yields an out-of-the-box calibrated network with competitive accuracy and robustness.
Study calibrates high-dimensional binary classifiers using angle between estimator and true weights.
problem Calibrating high-dimensional binary classifiers with provable properties.
method Interpolates with a chance classifier to construct well-calibrated predictor based on angle between estimator and true weights.
result Angular calibration approach is provably well-calibrated in high dimensions, minimizing Bregman divergence.
CalArena benchmarks post-hoc calibration methods across various tasks.
problem Inconsistent evaluations of post-hoc calibration methods.
method Large-scale benchmark with 2000 experiments, covering diverse models and settings.
result Smooth calibration functions outperform binning-based approaches.
Enhances out-of-domain calibration of neural networks.
problem Improving calibration performance of deep neural networks in out-of-domain settings.
method Consistency-guided temperature scaling (CTS) that considers style and content consistency.
result Significantly enhances out-of-domain calibration performance.
Temperature scaling fails for distributions with class overlaps, while Mixup improves calibration.
problem Temperature scaling's performance degrades with class overlaps, leading to poor calibration.
method Identified temperature scaling's limitations and compared it with Mixup for calibration.
result Mixup significantly outperforms temperature scaling in calibration metrics with class overlaps.
ATS improves calibration of deep networks, especially for small validation sets and noisy labels.
problem Improper calibration of deep neural networks, especially in small validation sets and noisy-labeled samples.
method Proposes Attended Temperature Scaling (ATS) to improve calibration of deep networks.
result ATS improves calibration of deep networks, especially for small validation sets and noisy-labeled samples.
The paper calibrates uncertainty in dropout variational inference models.
problem Miscalibration of model uncertainty in dropout variational inference.
method Logit scaling methods are extended to recalibrate model uncertainty.
result Logit scaling reduces miscalibration, improving reliability of predictions.
Calibrated PRMs improve inference efficiency for LLMs by dynamically adjusting compute budgets.
problem Poor calibration of PRMs leads to overestimation of success probabilities in partial reasoning steps.
method Quantile regression for calibration, instance-adaptive scaling (IAS) framework.
result Calibrated PRMs reduce inference costs while maintaining accuracy, especially on confident problems.
Study compares various calibration methods for binary classification tasks.
problem Improving probabilistic predictions in binary classification models.
method Benchmarked 21 classifiers using 5 calibration methods on real data.
result Venn-Abers predictors and Beta calibration show the largest log-loss reductions.
Proposes FIPO-BC for efficient online calibration of complex models.
problem Efficiently calibrating computationally expensive models with large datasets.
method Fixed inducing points online Bayesian calibration (FIPO-BC) algorithm.
result FIPO-BC is at least ten times faster than standard methods and enables online updates.
Online Platt Scaling adapts to varying data distributions.
problem Adapting Platt scaling to non-i.i.d. settings with distribution drift.
method Combines Platt scaling with online logistic regression and calibeating.
result OPS+calibeating method is guaranteed to be calibrated for adversarial outcomes.
New method calibrates classifier probabilities with guaranteed coverage.
problem Inaccurate probability estimates by classifiers in high-risk applications.
method Adaptive temperature scaling algorithm for conformal prediction.
result Improves calibration error measures and standard metrics across various tasks.
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.
The paper studies entropy calibration in language models and finds that miscalibration improves slowly with scale.
problem The problem is whether language model entropy calibration improves with scale and if it's possible to calibrate without reducing log loss.
method The authors study a simplified theoretical setting to characterize miscalibration scaling behavior and measure it empirically in language models ranging from 0.5B to 70B parameters.
result The observed scaling behavior of miscalibration is similar to theoretical predictions, indicating slow improvement with scale. The authors also prove theoretically that it is possible to reduce entropy while preserving log loss if access to a black box predicting future entropy is available.
Paper develops a new method to improve model calibration under distribution shifts.
problem Challenges in uncertainty quantification with different training and test distributions.
method Develops multi-domain temperature scaling to handle distribution shifts.
result Outperforms existing methods on in-distribution and out-of-distribution test sets.
Proposes a new method to improve multiclass probability calibration.
problem Uncalibrated class probabilities in multiclass classifiers leading to over-confidence.
method Dirichlet calibration method applicable to any model class, derived from Dirichlet distributions.
result Improved probabilistic predictions across various datasets and classifiers.
Novel framework for unbiased confidence estimates in object detection.
problem Unbiased confidence estimates for safety-critical object detection.
method Combines regression output with additional information for calibration.
result Calibrated confidence estimates for image location and scale.
New method calibrates noise for attack risk, improving ML model accuracy.
problem Improving accuracy of privacy-preserving ML models while maintaining privacy.
method Directly calibrates noise scale to a desired attack risk level, bypassing the standard ε \varepsilon ε -calibration. result Significantly decreases noise scale, leading to increased utility at the same risk level.
UTS improves DNN uncertainty calibration without labels, robust to domain shift.
problem Improving uncertainty calibration of DNNs under domain shift.
method UTS uses unlabeled test samples and a novel weighted NLL loss function.
result UTS outperforms other methods in domain shift scenarios.
New bin-wise scaling methods improve prediction uncertainty calibration for machine learning.
problem Improving prediction uncertainty calibration for machine learning regression.
method Adaptations of Binwise Variance Scaling (BVS) with alternative loss functions and feature-based binning.
result Improved adaptivity and consistency in prediction uncertainty calibration.
PD curve calibration refers to the transformation of a set of rating grade level probabilities of default (PDs) to another average PD level that is determined by a change of the underlying portfolio-wide PD. This paper presents a framework that allows to explore a variety of calibration approaches and the conditions un…
Focal loss improves deep neural networks' accuracy and calibration.
problem Miscalibration in deep neural networks.
method Using focal loss and temperature scaling to improve model calibration.
result Focal loss leads to state-of-the-art calibrated models without sacrificing accuracy.
Temperature scaling improves model uncertainty but not diversity in LLMs.
problem Improving the calibration and stochasticity of probabilistic models.
method Investigates theoretical properties of temperature scaling in classification and LLMs.
result Temperature scaling increases model uncertainty but not diversity in LLMs.
Annealing Double-Head calibrates deep neural networks during training.
problem Overestimation or underestimation of predictive confidence in deep neural networks.
method An additional calibration head and Annealing technique to dynamically scale logits.
result State-of-the-art model calibration performance achieved without post-processing.
Calibration without labels in multiple testing
problem Interpretable error probabilities in large-scale hypothesis testing
method Constructing pseudo-labels from spacings of ordered p p p -values result Finding that q q q -value can be severely miscalibrated Obtaining accurate and well calibrated probability estimates from classifiers is useful in many applications, for example, when minimising the expected cost of classifications. Existing methods of calibrating probability estimates are applied globally, ignoring the potential for improvements by applying a more fine-gra…
A new calibration metric bridges testability and actionability.
problem Combining testability and actionable insights for forecast probabilities.
method Cutoff Calibration Error (CCE) that assesses calibration over intervals of forecasted probabilities.
result Cutoff Calibration Error is both testable and actionable.
This work addresses building fair and calibrated models.
problem Building models that are both fair and calibrated.
method Developed a new definition of fairness and showed that group-wise calibration results in fairness. Proposed post-processing techniques and modifications of calibration losses.
result Demonstrated that ensuring group-wise calibration results in a fair model under the new definition of fairness.
CAIRO separates ranking from scaling to improve robustness.
problem Conflating ranking and scaling in regression leads to model vulnerability.
method Two-stage approach: first learns a scoring function, then recovers scale.
result CAIRO recovers true regression function with auto-calibration guarantees.
New calibration measures for multi-class classification improve model accuracy.
problem Calibration of multi-class classification models is insufficient for safety-critical applications.
method Developed new calibration measures and estimators for multi-class classification.
result Proposed estimators improve interpretability and accuracy of calibration measures.
Boosted decision trees typically yield good accuracy, precision, and ROC area. However, because the outputs from boosting are not well calibrated posterior probabilities, boosting yields poor squared error and cross-entropy. We empirically demonstrate why AdaBoost predicts distorted probabilities and examine three cali…
TS improves class coverage but reduces CP set size, offering a trade-off.
problem Combining temperature scaling with conformal prediction for deep classifiers.
method Empirical study and mathematical theory of TS's effect on CP.
result TS allows trading prediction set size and conditional coverage.
This paper investigates uncertainty calibration in multimodal large language models.
problem Challenges in properly calibrating uncertainty in multimodal large language models.
method Investigation of representative MLLMs across various scenarios, including visual fine-tuning and multimodal training.
result MLLMs tend to give answers rather than admit uncertainty, but this self-assessment improves with proper prompt adjustments.
New method for high-dimensional regression with unknown scale parameter.
problem High-dimensional linear regression with unknown scale parameter.
method Penalized Huber M M M -estimator with adaptive Lepski's method. result The method effectively calibrates scale in high-dimensional robust regression.
Paper proposes a method to predict MOBA game winners with calibrated confidence.
problem Predicting MOBA game winners with noisy data and uncertain noise.
method A novel confidence-calibration method considering data uncertainty.
result Achieves outstanding expected calibration error (ECE) of 0.57%.
This paper improves Reifenberg's theorem for almost calibrated sets, ensuring rectifiability with volume bounds.
problem Improving the rectifiability of sets that are close to subspaces under certain calibrations.
method Using ε-calibrations and positivity conditions, the paper shows that almost calibrated sets are rectifiable with volume bounds.
result Almost calibrated sets are rectifiable with uniform volume bounds.
Proposes top-label calibration and M2B framework for multiclass to binary calibration.
problem Multiclass calibration and interpretation issues.
method Top-label calibration and M2B reduction framework.
result M2B + HB achieves lower calibration error than other methods.
Improved online algorithm for convex losses with near-optimal swap regret.
problem Minimizing swap regret for convex losses.
method Randomized online algorithm using multi-scale binning.
result Near-optimal O ~ ( T ) \widetilde O(\sqrt T) O ( T ) expected swap regret. CRUDE calibrates regression uncertainty without assuming specific error distributions.
problem Uncalibrated uncertainty estimates in regression models, especially for modern predictive tasks.
method CRUDE assumes error distributions have a constant shape, shifted and scaled by predicted mean and standard deviation.
result CRUDE produces sharper, better calibrated, and more accurate uncertainty estimates than existing methods.
Study explores calibration properties in neural architectures.
problem Calibration issues in deep neural networks despite improved accuracy.
method Leverages Neural Architecture Search (NAS) to evaluate 117,702 neural networks.
result Identifies key architectural designs beneficial for calibration.
Paper proposes ensemble distillation for well-calibrated structured prediction.
problem Well-calibrated predictions are hard to achieve in structured prediction.
method Ensemble distillation framework for structured prediction.
result Ensemble distillation produces well-calibrated models with similar performance and calibration benefits to ensembles.
Bayesian neural networks outperform calibrated neural networks for tabular data.
problem Uncertainty in neural network predictions for tabular data.
method Bayesian neural networks vs. post-hoc calibration methods.
result Bayesian neural networks yield competitive performance compared to calibrated neural networks.
New CAGP-GS framework improves Gaussian process uncertainty quantification.
problem Scaling issue in Gaussian processes for large datasets.
method Calibrated probabilistic linear solvers for reduced complexity.
result CAGP-GS framework provides more realistic uncertainty quantification.
New method calibrates deep networks by preserving top-k predictions.
problem Calibrated confidence scores for multi-class deep networks to avoid rare mistakes.
method Intra order-preserving functions combined with neural network architecture.
result Outperforms state-of-the-art methods in evaluation metrics.
SGPA calibrates transformer uncertainty for safety-critical tasks.
problem Uncertainty estimation in transformer models for safety-critical domains.
method Bayesian inference in transformer's output space using sparse Gaussian processes.
result SGPA-based Transformers improve in-distribution calibration and out-of-distribution robustness.
Optimizing full likelihoods adapts loss scales and shapes for robust modeling.
problem Rigid loss functions limit model adaptability and robustness.
method Optimize full likelihoods with adjustable parameters.
result Adaptive tuning of loss scales and shapes improves model robustness.