awesome-confidence-calibration

• [ICML 2017] On Calibration of Modern Neural Networks paper

• [ALMC 1997] Probabilistic outputs for support vector machines and comparison to regularized likelihood methods paper

• [ICLR 2020] distance-based learning from errors for confidence calibration paper

• [Arxiv 2019] Confidence Calibration for Convolutional Neural Networks Using Structured Dropout paper

• [ICML 2016] Dropout as a bayesian approximation: Representing model uncertainty in deep learning paper code

• [NIPS 2020] Improving model calibration with accuracy versus uncertainty optimization paper code

• [AISTATS 2011] Approximate inference for the loss-calibrated Bayesian paper

• [Arxiv 2018] Loss-calibrated approximate inference in bayesian neural networks paper

• [CVPR 2019] Learning for Single-Shot Confidence Calibration in Deep Neural Networks through Stochastic Inferences paper

• [ICML 2005] Predicting Good Probabilities With Supervised Learning apper

• [KDD 2002] Transforming Classifier Scores into Accurate Multiclass Probability Estimates paper

• [NIPS 2017] On Fairness and Calibration paper

• [Arxiv 2016] Approximating Likelihood Ratios with Calibrated Discriminative Classifiers paper

• [NIPS 2020] Calibrating Deep Neural Networks using Focal Loss paper

• [NIPS 2019] Verified Uncertainty Calibration paper code

• [CVPR 2021] Improving Calibration for Long-Tailed Recognition paper code

• [CVPR 2021] Post-hoc Uncertainty Calibration for Domain Drift Scenarios paper code

• [NIPS 2017] Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles apper

• [NIPS 2019] Addressing Failure Prediction by Learning Model Confidence paper code

• [ICML 2018] Accurate Uncertainties for Deep Learning Using Calibrated Regression paper

• [ICLR 2018] Training Confidence-calibrated Classifiers for Detecting Out-of-Distribution Samples paper code

• [NIPS 2019] Addressing Failure Prediction by Learning Model Confidence paper code

• [Arxiv 2021] On the Calibration and Uncertainty of Neural Learning to Rank Models paper code

• [Classic] Isotonic Regression paper paper paper

• [NIPS 2019] Beyond temperature scaling: Obtaining well-calibrated multi-class probabilities with Dirichlet calibration paper

• [ICML 2001] Obtaining calibrated probability estimates from decision trees and naive bayesian classifiers paper

• [AISTATS 2017] Beyond sigmoids: How to obtain well-calibrated probabilities from binary classifiers with beta calibration paper code

• [Arxiv 2015] Binary classifier calibration using an ensemble of near isotonic regression models. paper

• [KDD 2019] Non-parametric Bayesian Isotonic Calibration: Fighting Over-Confidence in Binary Classification paper code

• [AAAI 2015] Obtaining Well Calibrated Probabilities Using Bayesian Binning paper code

• [Arxiv 2021] Distribution-free calibration guarantees for histogram binning without sample splitting paper code

• [ICML 2012] Predicting accurate probabilities with a ranking loss paper

Resources

• Statistical Decision Theory and Bayesian Analysis book

• A Tutorial on Learning With Bayesian Networks paper

Application

• [AAAI 2021] Learning to Cascade: Confidence Calibration for Improving the Accuracy and Computational Cost of Cascade Inference Systems paper