Target Strangeness: A Novel Conformal Prediction Difficulty Estimator
This work addresses a specific challenge in conformal prediction for researchers and practitioners, offering an incremental improvement over existing methods.
The paper tackles the problem of estimating difficulty in conformal prediction by introducing Target Strangeness, a novel estimator that normalizes prediction intervals based on atypical predictions among nearest neighbors, achieving state-of-the-art performance in conformal regression experiments.
This paper introduces Target Strangeness, a novel difficulty estimator for conformal prediction (CP) that offers an alternative approach for normalizing prediction intervals (PIs). By assessing how atypical a prediction is within the context of its nearest neighbours' target distribution, Target Strangeness can surpass the current state-of-the-art performance. This novel difficulty estimator is evaluated against others in the context of several conformal regression experiments.