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.
The paper addresses calibration in label ranking, a structured prediction task.
problem Calibration in label ranking is not well understood and often poorly calibrated.
method Formalized calibration for label ranking, developed a hierarchy of notions, and empirically evaluated models.
result Popular label ranking models are often poorly calibrated, with differences between sub-ranking and top-k metrics.
Estimates calibration error under label shift without labels.
problem Ensuring model reliability in the face of dataset shift without access to labels.
method Importance re-weighting of the labeled source distribution to estimate calibration error under label shift.
result Effective and reliable CE estimation with respect to the shifted target distribution.
Proposes efficient calibration for indoor localization models.
problem Calibration data scarcity in wireless indoor localization.
method Uses synthetic labels and prediction sets to fine-tune a predictor and estimate bias.
result Yields rigorous coverage guarantees for prediction sets.
Improves confidence calibration in neural networks by smoothing labels based on class similarity.
problem Improving confidence calibration in deep neural networks for safety-critical applications.
method Proposes a novel label smoothing technique where label values are based on similarities with the reference class, using different similarity measurements.
result Consistently outperforms state-of-the-art calibration techniques on various datasets and network architectures.
Paper estimates optimal classification error with soft labels and calibration.
problem Estimating the optimal classification error with soft labels and calibration.
method Extends previous work on soft labels to estimate Bayes error, addressing bias and corrupted labels.
result The method provides a statistically consistent estimator of the Bayes error, even with imperfectly calibrated soft labels.
New method calibrates multi-class predictions efficiently without sacrificing accuracy.
problem Efficiently calibrating multi-class predictions without sacrificing accuracy.
method Formulated robust projected smooth calibration and new recalibration algorithms.
result Achieves strong guarantees for binary classification tasks with polynomial complexity.
Label shift refers to the phenomenon where the prior class probability p(y) changes between the training and test distributions, while the conditional probability p(x|y) stays fixed. Label shift arises in settings like medical diagnosis, where a classifier trained to predict disease given symptoms must be adapted to sc…
Paper presents unsupervised calibration for split conformal classification.
problem Inconvenient requirement of labeled calibration samples.
method Uses unsupervised calibration samples alongside supervised training samples.
result Achieves comparable performance to supervised calibration methods.
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 Conformal Bayes under label shift: post-hoc calibration vs. in-training adaptation
problem Bayesian prediction sets under label shift
method Post-hoc calibration vs. In-training adaptation
result Both strategies achieve valid coverage equally in an unbiased training regime
The study examines when to trust confidence thresholding in pseudo-labelling regression.
problem Calibrated probabilities from classifiers used for pseudo-labelling need careful handling to avoid bias in downstream regression.
method Developed a diagnostic apparatus to predict and bound the bias induced by confidence thresholding, derived a closed-form expression for the attenuation bias.
result The bias can be predicted from the residual score variance V ∗ V^{*} V ∗ , motivating a structural separation between classifier features and downstream controls. CAT improves domain adaptation for fault diagnosis by calibrating teacher network predictions.
problem Performance drops in deep learning models when applied to different data distributions.
method CAT uses post-hoc calibration techniques to calibrate predictions of the teacher network during self-training.
result CAT achieves state-of-the-art performance on most transfer tasks in domain-adaptive IFD.
TCR improves DNN robustness to noisy labels with minimal overhead.
problem Training on noisy labeled datasets degrades DNN generalization.
method TCR combines original labels and previous epoch predictions for regularization.
result TCR consistently enhances DNN robustness to label noise.
Improves reliability of medical diagnosis uncertainty estimates.
problem Label uncertainty in medical diagnosis.
method Post-hoc a l p h a alpha a l p ha -calibration method for neural network classifiers. result Significantly enhances reliability of uncertainty estimates.
Improved segmentation model adaptation for new domains.
problem Reduced performance of pre-trained models on new domains.
method Calculated soft-label prototypes and predicted closest to class probabilities.
result Significant performance improvements on synthetic-to-real segmentation.
The study examines label smoothing to improve confidence calibration in fine-tuned LLMs.
problem Improving confidence calibration in fine-tuned large language models (LLMs) after instruction tuning.
method Examine various open-sourced LLMs, label smoothing, and custom kernel design.
result Label smoothing is effective in maintaining confidence calibration but faces challenges in large vocabulary LLMs.
Noise-Aware Conformal Prediction (NACP) calibrates CP for noisy labels.
problem Calibrating Conformal Prediction with noisy labels.
method Estimate conformal threshold from noisy labels using uniform noise coverage guarantee.
result Finite sample coverage guarantee for uniform noise remains effective in high-class tasks.
Improves model calibration by addressing adversarially unrobust inputs.
problem Miscalibrated predictions and lack of adversarial robustness in neural networks.
method AR-AdaLS, an adaptive label smoothing method that considers adversarial robustness.
result AR-AdaLS improves model calibration, even under distributional shifts.
ELSA efficiently adapts to label shift without post-prediction calibrations.
problem Domain adaptation with label shift across training and testing datasets.
method Moment-matching framework based on influence function geometry; solves linear systems for adaptation weights.
result ELSA estimator is n \sqrt{n} n -consistent and asymptotically normal, achieving state-of-the-art estimation performance. MEC improves efficiency and robustness in semi-supervised inference.
problem Efficient inference with limited labeled data and robust uncertainty quantification.
method Machine-Learning-Assisted Generalized Entropy Calibration (MEC) using cross-fitted, calibration-weighted PPI.
result MEC achieves semiparametric efficiency bounds under weaker assumptions and provides near-nominal coverage.
Two approaches improve conformal Bayes for label shift, one post-hoc and one in-training.
problem Improving prediction sets for target domain under label shift.
method Two complementary approaches: post-hoc calibration and in-training adaptation.
result In-training adaptation achieves up to 43% width reduction at unchanged coverage.
New method improves auto-labeling accuracy by optimizing confidence functions.
problem Overconfident model scores lead to poor TBAL performance.
method Developed a new post-hoc method, Colander, to optimize TBAL confidence functions.
result Achieves up to 60% improvement in coverage over baseline methods.
Fairness audits fail under missing protected labels, especially at zero access.
problem Understanding the reliability of fairness audits with incomplete protected-label data.
method Introduced a seed-calibrated stress test to separate missingness effects from seed-to-seed movement.
result Missing protected labels do not significantly alter fairness mitigation methods, but they can lead to harmful intersectional outcomes.
Label smoothing improves model performance even with noisy labels.
problem Mitigating label noise in deep learning models.
method Examined label smoothing as a technique to cope with label noise and compared it to loss-correction methods.
result Label smoothing is competitive with loss-correction techniques under label noise and beneficial for distillation from noisy data.
Advances rule-based multi-label classification using conformal prediction.
problem Improving accuracy and decision making in multi-label classification.
method Combines conformal prediction with rule-based learning to provide natural conformity scores and calibrate rule assessments.
result Calibrated conformity scores enhance prediction accuracy and decision making.
The paper investigates how dataset quality and heterogeneity affect model confidence in machine learning.
problem Understanding how dataset quality and heterogeneity impact model confidence in machine learning.
method The study uses theoretical explanations and experimental demonstrations to investigate the effects of dataset size, label noise, and class heterogeneity on model confidence.
result Label noise reduces model confidence, while reduced dataset size increases it, and class heterogeneity leads to inconsistent confidence across classes.
Unified view of label shift estimation methods.
problem Label distribution changes but class-conditional distributions remain the same.
method Unified view of two approaches: BBSE and MLLS.
result Unified framework and theoretical characterization of MLLS.
Study shows label errors impact model disparity metrics, proposing mitigation methods.
problem Impact of label errors on model disparity metrics.
method Empirical study, characterizing label error effects; proposing estimation and relabeling methods.
result Label errors significantly affect model disparity metrics, particularly for minority groups.
Proposes a new method for deep ensembles that improves accuracy and calibration.
problem Improving accuracy and calibration of deep ensembles.
method Estimates confusion matrices of ensemble members and weighs them according to their inferred performance.
result Empirically shows superiority of soft Dawid Skene over ensemble averaging.
New truthful calibration errors improve model ranking in multiclass prediction.
problem Non-truthful calibration errors can mislead model comparisons.
method Introduced perfectly truthful calibration errors for multiclass predictions.
result Truthful calibration errors preserve decision-theoretic dominance and stabilize model rankings.
Label noise in adversarial training leads to robust overfitting, explained and mitigated.
problem Label noise in adversarial training causes robust overfitting.
method Proposed a method to automatically calibrate labels.
result Consistent performance improvements across various models and datasets.
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.
Trained DNN models are increasingly adopted as integral parts of software systems, but they often perform deficiently in the field. A particularly damaging problem is that DNN models often give false predictions with high confidence, due to the unavoidable slight divergences between operation data and training data. To…
Calibrates network confidence for unsupervised domain adaptation.
problem Calibrating a model trained on a source domain to a target domain without labeled data.
method Estimates network accuracy on the target domain and calibrates prediction confidence directly in the target domain.
result Significantly outperforms existing methods across standard datasets.
E-Stitchup augments pre-trained embeddings for better model performance and confidence calibration.
problem Improving classification accuracy and confidence calibration of models using pre-trained embeddings.
method Data augmentation methods inspired by Mixup combined with label softening.
result Significantly increased classification accuracy and reduced training time.
Framework calibrates ML models for risk control in various tasks.
problem Achieving statistical guarantees for model predictions.
method Reframing risk control as multiple hypothesis testing, applying statistical techniques.
result New calibration methods for multi-label classification, instance segmentation, outlier detection, and confidence set coverage.
TransCal calibrates DA models with lower bias and variance.
problem Calibrating DA models to estimate accurate predictive uncertainty.
method Transferable Calibration (TransCal) in a unified hyperparameter-free optimization framework.
result TransCal achieves more accurate calibration with lower bias and variance.
AdapTable adapts tabular models to shifts without source data, improving HELOC performance.
problem Distribution shifts in tabular data threaten model performance.
method Shift-aware uncertainty calibrator and label distribution handler.
result Up to 16% improvement on HELOC dataset.
The paper tackles uncertainty quantification for classification under label shift without assuming i.i.d. data.
problem Uncertainty quantification for classification under label shift in non-i.i.d. settings.
method The paper uses conformal prediction and post-hoc binning for distribution-free UQ, and reweights these methods for label shift.
result The reweighted methods improve UQ performance under label shift, preserving coverage and calibration.
Excellent ranking power along with well calibrated probability estimates are needed in many classification tasks. In this paper, we introduce a technique, Calibrated Boosting-Forest that captures both. This novel technique is an ensemble of gradient boosting machines that can support both continuous and binary labels. …
Proposes a transductive matrix completion method with calibration for multi-task learning.
problem Improving multi-task learning with multiple related data sources.
method Transductive matrix completion with calibration constraint.
result The proposed algorithm recovers incomplete feature and target matrices with improved results.
Adapts self-supervised learning using probabilistic sets with validity guarantees.
problem Lack of validity guarantees in pseudo-labels from self-supervised learning.
method Uses conformal prediction to provide validity guarantees for probabilistic labels.
result Valid probabilistic labels improve calibration and performance.
We present surrogate regret bounds for arbitrary surrogate losses in the context of binary classification with label-dependent costs. Such bounds relate a classifier's risk, assessed with respect to a surrogate loss, to its cost-sensitive classification risk. Two approaches to surrogate regret bounds are developed. The…
Study shows multilingual LLM calibration effects improve model confidence but not accuracy.
problem Improving multilingual language model calibration in low-resource settings.
method Analysis of two multilingual benchmarks using instruction-tuning and label smoothing.
result Model confidence increases in low-resource languages after instruction-tuning but accuracy improvements are marginal.
Method provides formal guarantees for decomposing model uncertainty.
problem Decomposing model uncertainty into aleatoric and epistemic components.
method Higher-order calibration using k-snapshots.
result Formal guarantees for aleatoric uncertainty matching real-world distribution.
Unsupervised domain adaptation (UDA) aims at inferring class labels for unlabeled target domain given a related labeled source dataset. Intuitively, a model trained on source domain normally produces higher uncertainties for unseen data. In this work, we build on this assumption and propose to adapt from source to targ…
EC method calibrates neural networks by matching average confidence to correct label proportion.
problem Overoptimism in neural network prediction confidence.
method Expectation consistency (EC) post-training rescaling of weights.
result EC achieves similar calibration performance to temperature scaling (TS) but is based on a principled Bayesian principle.