Conformal Prediction for Reliable Machine Learning

1.1 The Basic Setting and Assumptions . . . . . . . . . . . . . . . . . . . . . . 3 1.2 Set and Confidence Predictors . . . . . . . . . . . . . . . . . . . . . . . . 4 1.2.1 Validity and Efficiency of Set and Confidence Predictors . . . . . . 5 1.3 Conformal Prediction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 1.3.1 The Binary Case . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.3.2 The Gaussian Case . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.4 Efficiency in the Case of Prediction without Objects . . . . . . . . . . . . . 9 1.5 Universality of Conformal Predictors . . . . . . . . . . . . . . . . . . . . . 11 1.6 Structured Case and Classification . . . . . . . . . . . . . . . . . . . . . . 13 1.7 Regression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 1.8 Additional Properties of Validity and Efficiency in the Online Framework . . . 15 1.8.1 Asymptotically Efficient Conformal Predictors . . . . . . . . . . . 17 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19