Conformal Prediction via Regression-as-Classification
This incremental work addresses unstable prediction intervals in regression for practitioners using conformal prediction.
The paper tackled the challenge of conformal prediction for regression with complex output distributions by converting regression to a classification problem and adapting techniques to preserve ordering, resulting in good performance on benchmarks.
Conformal prediction (CP) for regression can be challenging, especially when the output distribution is heteroscedastic, multimodal, or skewed. Some of the issues can be addressed by estimating a distribution over the output, but in reality, such approaches can be sensitive to estimation error and yield unstable intervals.~Here, we circumvent the challenges by converting regression to a classification problem and then use CP for classification to obtain CP sets for regression.~To preserve the ordering of the continuous-output space, we design a new loss function and make necessary modifications to the CP classification techniques.~Empirical results on many benchmarks shows that this simple approach gives surprisingly good results on many practical problems.