paper-with-me

홈 › Papers

Calibration by Distribution Matching: Trainable Kernel Calibration Metrics

2023-10-31 · NeurIPS 2023 11 · Charles Marx, Sofian Zalouk, Stefano Ermon

Calibration ensures that probabilistic forecasts meaningfully capture uncertainty by requiring that predicted probabilities align with empirical frequencies. However, many existing calibration methods are specialized for post-hoc recalibration, which can worsen the sharpness of forecasts. Drawing on the insight that calibration can be viewed as a distribution matching task, we introduce kernel-based calibration metrics that unify and generalize popular forms of calibration for both classification and regression. These metrics admit differentiable sample estimates, making it easy to incorporate a calibration objective into empirical risk minimization. Furthermore, we provide intuitive mechanisms to tailor calibration metrics to a decision task, and enforce accurate loss estimation and no regret decisions. Our empirical evaluation demonstrates that employing these metrics as regularizers enhances calibration, sharpness, and decision-making across a range of regression and classification tasks, outperforming methods relying solely on post-hoc recalibration.

📄 PDF Abstract BibTeX arXiv:2310.20211

Code (1)

kernel-calibration/kernel-calibration 공식 구현 pytorch

Tasks

Decision Makingregression

Methods 이 논문이 사용한 방법론

ALIGN In the ALIGN method, visual and language representations are jointly trained from noisy image alt-text data. The image and text encoders are learned via contrastive loss…

Similar Papers 제목 키워드 기반

Calibration Regularized Training of Deep Neural Networks using Kernel Density Estimation

2021-09-29 · Teodora Popordanoska, Raphael Sayer, Matthew B. Blaschko

Calibrated probabilistic classifiers are models whose predicted probabilities can directly be interpreted as uncertainty estimates. This property is particularly important in safety-critical applications such as medical …

Autonomous DrivingDensity EstimationMedical Diagnosis

Trainable Calibration Measures for Neural Networks from Kernel Mean Embeddings

2018-07-01 · ICML 2018 7 · Aviral Kumar, Sunita Sarawagi, Ujjwal Jain

Modern neural networks have recently been found to be poorly calibrated, primarily in the direction of over-confidence. Methods like entropy penalty and temperature smoothing improve calibration by clamping confiden…

A Consistent and Differentiable Lp Canonical Calibration Error Estimator

2022-10-13 · Teodora Popordanoska, Raphael Sayer, Matthew B. Blaschko

Calibrated probabilistic classifiers are models whose predicted probabilities can directly be interpreted as uncertainty estimates. It has been shown recently that deep neural networks are poorly calibrated and tend to o…

Simple Calibration via Geodesic Kernels

2022-01-31 · Jayanta Dey, Haoyin Xu, Ashwin De Silva, Joshua T. Vogelstein

Deep discriminative approaches, such as decision forests and deep neural networks, have recently found applications in many important real-world scenarios. However, deploying these learning algorithms in safety-critical …

Out-of-Distribution Detectionregression

All Models Are Miscalibrated, But Some Less So: Comparing Calibration with Conditional Mean Operators

2025-02-17 · Peter Moskvichev, Dino Sejdinovic

When working in a high-risk setting, having well calibrated probabilistic predictive models is a crucial requirement. However, estimators for calibration error are not always able to correctly distinguish which model is …

All