Proposes MCLLO for assessing and recalibrating multiclass probability predictions.
problem Limited multicategory recalibration methods for assessing and comparing model calibration.
method MCLLO recalibration method that assesses calibration without model access and is easy to interpret.
result MCLLO outperforms other methods in simulations and real-world case studies.
New framework for evaluating multiclass classifier calibration.
problem Ensuring classifiers are well-calibrated for trustworthy predictions.
method Utility Calibration framework that measures calibration error relative to a utility function.
result Unified and robust interpretation of existing calibration metrics.
This paper argues against using calibration metrics for assessing posterior probabilities and proposes expected proper scoring rules instead.
problem The assessment of posterior probabilities generated by machine learning classifiers using calibration metrics is flawed and should be replaced with expected proper scoring rules.
method The paper reviews proper scoring rules from a practical perspective, explains why expected PSRs are a principled measure of posterior quality, and introduces a new calibration metric called calibration loss.
result Calibration loss is superior to expected calibration error and expected score divergence calibration metrics for assessing posterior probabilities.
A framework assesses the trustworthiness of probabilistic classifiers using local calibration error.
problem Assessing the trustworthiness of probabilistic classifiers beyond traditional metrics.
method I-trustworthy framework linking local calibration to trustworthiness; Kernel Local Calibration Error (KLCE) method for hypothesis testing.
result The effectiveness of the proposed test statistic demonstrated through simulated and real-world datasets.
New test assesses probabilistic model calibration without expensive approximations.
problem Assessing calibration of probabilistic models with scores.
method Kernel Calibration Conditional Stein Discrepancy (KCCSD) test using new score-based kernels.
result Control over type-I error with improved scalability and efficiency.
Proposes CCE to assess point-wise reliability of neural network predictions.
problem Overconfidence and misaligned predictive distributions in neural networks.
method Introduces Conditional Congruence (CCE) metric using conditional kernel mean embeddings.
result CCE exhibits correctness, monotonicity, reliability, and robustness in high-dimensional regression tasks.
This guide clarifies techniques for assessing and comparing model calibration and performance.
problem Assessing and comparing the calibration and performance of predictive models in insurance and actuarial practice.
method Clarifies statistical techniques for assessing model calibration and comparing models, emphasizing the importance of specifying the prediction target functional and choosing the appropriate scoring function.
result Provides guidance for the practical choice of scoring functions and illustrates results with real data case studies.
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.
LLMs show surprising confidence in their answers, beyond just tokens.
problem LLMs lack meaningful confidence estimates for their responses.
method Semantic calibration test based on local loss optimality and equivalence classes.
result Base LLMs are semantically calibrated across tasks, contrary to expectations.
Calibration without labels in multiple testing
problem Interpretable error probabilities in large-scale hypothesis testing
method Constructing pseudo-labels from spacings of ordered p-values result Finding that q-value can be severely miscalibrated New framework assesses LM uncertainty without thresholding.
problem Uncertainty quantification for LMs, especially comparing different measures.
method Rank-Calibration framework to assess uncertainty and confidence measures.
result Higher uncertainty correlates with lower generation quality.
This paper reviews metrics to assess AI model calibration accuracy.
problem AI model probabilities do not always match their true accuracy.
method Comprehensive review of 82 probability calibration metrics.
result Identified 4 classifier families and 1 object detection family of metrics.
Paper proposes a natural hedging framework with graphical assessment for longevity risk management.
problem Lack of a unified framework for natural hedging and graphical risk assessment.
method Structured natural hedging framework integrated with a graphical risk metric.
result Demonstrates flexibility, interpretability, and practical value for longevity risk management.
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.
TraCeR uses transformers to analyze survival data with longitudinal covariates.
problem Handling longitudinal covariates and assessing model calibration in survival analysis.
method Transformer-based survival analysis framework with factorized self-attention architecture.
result TraCeR achieves significant performance improvements over state-of-the-art methods.
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.
New methods reduce bias in estimating calibration error.
problem Reducing bias in estimating calibration error.
method Synthesizing model outputs and using equal-mass bins.
result Two reliable calibration-error estimators found: debiased estimator and ECE_sweep.
A method to monitor probability predictions for calibration loss in image classification models.
problem Maintaining calibration in machine learning predictions over time.
method Cumulative sum-based approach with dynamic limits for early detection of miscalibration.
result Early detection of operational context changes impacting image classification performance.
Efficiently calibrates Bergomi models to VIX derivatives using vector quantization.
problem Calibrating Bergomi models to VIX derivatives for accurate pricing.
method Applied vector quantization in mixed Bergomi models for fast and efficient option pricing.
result Calibration of Bergomi models to VIX derivatives is feasible and accurate over daily data.
Study shows non-systematic bias in customer satisfaction surveys limits data value.
problem Non-systematic bias in customer satisfaction surveys limits data value.
method Used real customer satisfaction survey data of a large retail bank to show the irreducible error and suggest thoughtful survey design methods.
result A thoughtful survey design can reduce non-systematic error in customer satisfaction surveys.
Time series foundation models are well-calibrated, improving over baseline models.
problem Calibration of time series foundation models for practical applications.
method Systematic evaluations of five time series foundation models and two baselines, assessing calibration, prediction heads, and long-term forecasting.
result Time series foundation models are consistently better calibrated than baseline models and do not show over- or under-confidence.
We extend a variational framework to estimate calibration errors for Lp divergences.
problem Ensuring predicted probabilities match observed class frequencies in machine learning.
method Extend variational framework to Lp divergences, separating over- and under-confidence. result Avoids overestimation and separates over- and under-confidence.
The paper explores local-correlation models for pricing complex financial contracts.
problem Calibrating synthetic quanto forward contracts and composite options.
method Design on-line calibration procedures for local and stochastic volatility models.
result Calibration performance of local-correlation models compared to simpler approximations.
Paper introduces a method to improve survival model calibration without sacrificing discrimination.
problem Survival models struggle to balance accurate ranking and event prediction.
method Uses conformal regression to enhance calibration without compromising discrimination.
result The approach improves model calibration across 11 real-world datasets.
Benchmark assesses fairness in algorithmic uncertainty, revealing consistent and calibrated estimates improve fairness.
problem Challenges in managing uncertainty in fairness evaluations for predictive algorithms.
method Introduces FairlyUncertain, an axiomatic benchmark for evaluating uncertainty in fairness.
result Consistent and calibrated uncertainty estimates improve fairness without explicit fairness interventions.
Boosting trees can test necessary conditions for regression model calibration.
problem Testing calibration and auto-calibration in regression models.
method Using boosting trees to test calibration and auto-calibration.
result Boosting trees prove to be very powerful in testing calibration and auto-calibration in large insurance datasets.
Paper assesses how pandemic data impacts mortality models.
problem Impact of pandemic data on mortality projections.
method Calibrated Li & Lee model with transformed weekly data.
result Impact quantified, scenarios generated for future mortality.
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.
Unified model for equity option pricing and interest-rate risk assessment.
problem Pricing short and medium-term equity options and interest-rate risk.
method Developed a stochastic modeling framework using Heston, Bates, and CIR models, calibrated using Fourier inversion and FFT.
result Calibration stability and convergence of parameter sets across models.
For many applications it is critical to know the uncertainty of a neural network's predictions. While a variety of neural network parameter estimation methods have been proposed for uncertainty estimation, they have not been rigorously compared across uncertainty measures. We assess four of these parameter estimation m…
Post-calibration improves the accuracy of causal effect estimation.
problem Improperly calibrated propensity scores lead to inaccurate causal effect estimation.
method Performed a simulation study to assess the impact of post-calibration on causal effect estimation.
result Post-calibration reduces the error in estimating the average treatment effect, especially for expressive uncalibrated statistical estimators.
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.
New method calibrates heterogeneous treatment effect models.
problem Difficulty in estimating and calibrating heterogeneous treatment effects.
method Defined and proposed a robust estimator for HTE calibration, based on doubly robust treatment effect estimators.
result Proposed method evaluates calibration of learned HTE models, addressing overfitting and high-dimensionality.
Study on estimating conditional risk in machine learning.
problem Estimating expected loss of prediction models given input features.
method Analyzed in classification and regression settings, showing equivalence to standard regression. Developed theoretical insights and empirical validation.
result Conditional risk calibration is distinct from existing uncertainty quantification problems.
Calibrated probabilistic solvers improve accuracy of ODE estimates.
problem Uncertainty in probabilistic ODE solutions is not well-calibrated for adaptive step sizes.
method Introduce and assess several calibration methods for probabilistic ODE solvers.
result Calibration methods interact efficiently with adaptive step-size selection, improving posteriors.
Decision makers increasingly rely on algorithmic risk scores to determine access to binary treatments including bail, loans, and medical interventions. In these settings, we reconcile two fairness criteria that were previously shown to be in conflict: calibration and error rate equality. In particular, we derive necess…
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…
Histogram binning method proven with guarantees without splitting data.
problem Proving theoretical guarantees for histogram binning without sample splitting.
method Using Markov property of order statistics to prove calibration guarantees for original method.
result Proves histogram binning has strong calibration guarantees without sample splitting.
New methods for better uncertainty prediction in ML.
problem Insufficient calibration in machine learning regression.
method Conditional calibration with respect to input features (adaptivity).
result Consistency and adaptivity are complementary, and good consistency does not guarantee good adaptivity.
Network theory assesses systemic risk in the insurance sector.
problem Detecting critical insurance companies in systemic risk.
method Complex network approach with weighted effective resistance centrality.
result Identifies companies with significant influence on network robustness.
In this paper we study the probabilistic properties of the posteriors in a speech recognition system that uses a deep neural network (DNN) for acoustic modeling. We do this by reducing Kaldi's DNN shared pdf-id posteriors to phone likelihoods, and using test set forced alignments to evaluate these using a calibration s…
Develops a method to make predictions more informative without sacrificing accuracy.
problem Tension between well-calibrated and bold predictions.
method Bayesian model selection-based approach for assessing calibration and maximizing boldness.
result Maximally emboldening predictions while maintaining desired level of calibration.
Machine learning models deployed in real-world applications are often evaluated with precision-based metrics such as F1-score or AUC-PR (Area Under the Curve of Precision Recall). Heavily dependent on the class prior, such metrics make it difficult to interpret the variation of a model's performance over different subp…
Recently, Deep Neural Networks (DNNs) have been achieving impressive results on wide range of tasks. However, they suffer from being well-calibrated. In decision-making applications, such as autonomous driving or medical diagnosing, the confidence of deep networks plays an important role to bring the trust and reliabil…
T-Cal tests model calibration with a minimax optimal test.
problem Detecting mis-calibration of predictive models using a finite validation dataset.
method T-Cal is a minimax optimal test for calibration based on a debiased plug-in estimator of the ℓ2-Expected Calibration Error (ECE). result T-Cal is a practical tool for testing the calibration of probabilistic classification methods.
The article compares neural networks and logistic regression for credit scoring and introduces a new probability calibration technique.
problem Improving credit scoring accuracy using machine learning techniques.
method Comparison of logistic regression and neural networks, feature importance assessment, temporal feature inclusion, and SURE probability calibration.
result Neural networks can slightly improve credit scoring performance, and SURE calibration technique enhances probability calibration.
New methods correct bias in LLM-as-a-Judge evaluations, but reliability depends on judge quality and model calibration.
problem Systematic bias in LLM-as-a-Judge evaluations using naive estimators.
method Analytical results, simulations, and real-data case study to diagnose reliability of corrected estimates.
result Corrected estimates, especially shared-calibration comparisons, can be unreliable under certain conditions.
Study improves neural network calibration for drug discovery.
problem Improper calibration of neural network predictions in drug discovery.
method Compared different metrics for model hyperparameter tuning and proposed Bayesian Linear Probing (BLP) method.
result Bayesian Linear Probing (BLP) improves model calibration and accuracy.