English
Related papers

Related papers: Measuring Calibration in Deep Learning

200 papers

Calibration is a fundamental property of a good predictive model: it requires that the model predicts correctly in proportion to its confidence. Modern neural networks, however, provide no strong guarantees on their calibration -- and can…

Machine Learning · Computer Science 2022-10-07 A. Michael Carrell , Neil Mallinar , James Lucas , Preetum Nakkiran

To be considered reliable, a model must be calibrated so that its confidence in each decision closely reflects its true outcome. In this blogpost we'll take a look at the most commonly used definition for calibration and then dive into a…

Methodology · Statistics 2025-09-16 Maja Pavlovic

Large Language Models (LLMs) have demonstrated inherent calibration capabilities, where predicted probabilities align well with correctness, despite prior findings that deep neural networks are often overconfident. Recent studies have…

Machine Learning · Computer Science 2025-11-04 Abhinav Joshi , Areeb Ahmad , Ashutosh Modi

Despite the great success of state-of-the-art deep neural networks, several studies have reported models to be over-confident in predictions, indicating miscalibration. Label Smoothing has been proposed as a solution to the over-confidence…

Computer Vision and Pattern Recognition · Computer Science 2023-01-31 Shuang Ao , Stefan Rueger , Advaith Siddharthan

This paper explores the calibration of a classifier output score in binary classification problems. A calibrator is a function that maps the arbitrary classifier score, of a testing observation, onto $[0,1]$ to provide an estimate for the…

Machine Learning · Computer Science 2022-04-29 Waleed A. Yousef , Issa Traore , William Briguglio

In this work, we highlight and perform a comprehensive study on calibration attacks, a form of adversarial attacks that aim to trap victim models to be heavily miscalibrated without altering their predicted labels, hence endangering the…

Machine Learning · Computer Science 2024-12-03 Stephen Obadinma , Xiaodan Zhu , Hongyu Guo

For users to trust model predictions, they need to understand model outputs, particularly their confidence - calibration aims to adjust (calibrate) models' confidence to match expected accuracy. We argue that the traditional calibration…

Computation and Language · Computer Science 2022-10-25 Chenglei Si , Chen Zhao , Sewon Min , Jordan Boyd-Graber

AI generated predictions increasingly inform decision making in critical tasks, and therefore must be trustworthy. One widely used measure of trustworthiness is calibration, which requires that the predictions match the true frequencies and…

Machine Learning · Computer Science 2026-05-19 Konstantina Bairaktari , Lunjia Hu , Huy L. Nguyen , Jonathan Ullman

Ensuring that classifiers are well-calibrated, i.e., their predictions align with observed frequencies, is a minimal and fundamental requirement for classifiers to be viewed as trustworthy. Existing methods for assessing multiclass…

Machine Learning · Computer Science 2025-10-30 Mahmoud Hegazy , Michael I. Jordan , Aymeric Dieuleveut

Class imbalance poses a significant challenge in classification tasks, where traditional approaches often lead to biased models and unreliable predictions. Undersampling and oversampling techniques have been commonly employed to address…

Machine Learning · Computer Science 2025-10-22 Matt Clifford , Jonathan Erskine , Alexander Hepburn , Raúl Santos-Rodríguez , Dario Garcia-Garcia

Neural networks have proven successful at learning from complex data distributions by acting as universal function approximators. However, they are often overconfident in their predictions, which leads to inaccurate and miscalibrated…

Machine Learning · Computer Science 2021-02-23 Jeffrey Willette , Juho Lee , Sung Ju Hwang

In a well-calibrated risk prediction model, the average predicted probability is close to the true event rate for any given subgroup. Such models are reliable across heterogeneous populations and satisfy strong notions of algorithmic…

Machine Learning · Computer Science 2023-07-31 Jean Feng , Alexej Gossmann , Romain Pirracchio , Nicholas Petrick , Gene Pennello , Berkman Sahiner

While significant progress has been made in specifying neural networks capable of representing uncertainty, deep networks still often suffer from overconfidence and misaligned predictive distributions. Existing approaches for measuring this…

Machine Learning · Computer Science 2025-10-24 Spencer Young , Riley Sinema , Cole Edgren , Andrew Hall , Nathan Dong , Porter Jenkins

When adapting ICL with or without fine-tuning, we are curious about whether the instruction-tuned language model is able to achieve well-calibrated results without suffering from the problem of overconfidence (i.e., miscalibration)…

Computation and Language · Computer Science 2025-05-23 Chengzu Li , Han Zhou , Goran Glavaš , Anna Korhonen , Ivan Vulić

Many classification applications require accurate probability estimates in addition to good class separation but often classifiers are designed focusing only on the latter. Calibration is the process of improving probability estimates by…

Machine Learning · Computer Science 2020-01-31 Tuomo Alasalmi , Jaakko Suutala , Heli Koskimäki , Juha Röning

Calibration measures quantify how much a forecaster's predictions violates calibration, which requires that forecasts are unbiased conditioning on the forecasted probabilities. Two important desiderata for a calibration measure are its…

Machine Learning · Computer Science 2025-03-05 Mingda Qiao , Eric Zhao

Accurate calibration of probabilistic predictive models learned is critical for many practical prediction and decision-making tasks. There are two main categories of methods for building calibrated classifiers. One approach is to develop…

Machine Learning · Statistics 2014-01-16 Mahdi Pakdaman Naeini , Gregory F. Cooper , Milos Hauskrecht

Risk prediction models are increasingly used in healthcare to aid in clinical decision making. In most clinical contexts, model calibration (i.e., assessing the reliability of risk estimates) is critical. Data available for model…

Deep Learning methods are known to suffer from calibration issues: they typically produce over-confident estimates. These problems are exacerbated in the low data regime. Although the calibration of probabilistic models is well studied,…

Machine Learning · Statistics 2021-11-30 Rahul Rahaman , Alexandre H. Thiery

We prove a fundamental impossibility theorem: neural networks cannot simultaneously learn well-calibrated confidence estimates with meaningful diversity when trained using binary correct/incorrect supervision. Through rigorous mathematical…

Machine Learning · Computer Science 2025-09-19 Arjun S. Nair , Kristina P. Sinaga