How conformal prediction can be applied to multi-class classification problems
conformal prediction is a powerful framework that can be applied to multi-class classification problems. It provides a way to make predictions with a measure of certainty, which is particularly useful when dealing with multiple classes.
In the previous chapters, we have already looked at how conformal prediction assigns a p-value to each class for a given instance in the context of multi-class classification.
The p-value represents the confidence level of the prediction for that class. The higher the p-value, the more confident the model is that the instance belongs to that class.
The procedure for applying conformal prediction to multi-class classification is as follows:
- Calibration: A portion of the training data, known as the calibration set, is set aside. The model is trained on the remaining data.
- Prediction: For each class, the model predicts class scores. The conformity score...
