Guaranteed Coverage Prediction Intervals with Gaussian Process Regression
This addresses the issue of misleading uncertainty estimates in regression for practitioners relying on GPR, though it is incremental as it combines existing methods.
The paper tackles the problem of unreliable uncertainty estimates in Gaussian Process Regression (GPR) when models are misspecified, by introducing an extension using Conformal Prediction (CP) that guarantees prediction intervals with required coverage, such as 95%, even under misspecification, as demonstrated in experiments showing its superiority over existing methods.
Gaussian Process Regression (GPR) is a popular regression method, which unlike most Machine Learning techniques, provides estimates of uncertainty for its predictions. These uncertainty estimates however, are based on the assumption that the model is well-specified, an assumption that is violated in most practical applications, since the required knowledge is rarely available. As a result, the produced uncertainty estimates can become very misleading; for example the prediction intervals (PIs) produced for the 95% confidence level may cover much less than 95% of the true labels. To address this issue, this paper introduces an extension of GPR based on a Machine Learning framework called, Conformal Prediction (CP). This extension guarantees the production of PIs with the required coverage even when the model is completely misspecified. The proposed approach combines the advantages of GPR with the valid coverage guarantee of CP, while the performed experimental results demonstrate its superiority over existing methods.