Prediction intervals in supervised machine learning bound the region where the true outputs of new samples may fall. They are necessary in the task of separating reliable predictions of a trained model from near random guesses, minimizing the rate of false positives, and other problem-specific tasks in applied machine learning. Many real problems have heteroscedastic stochastic outputs, which explains the need of input-dependent prediction intervals. This paper proposes to estimate the input-dependent prediction intervals by a separate extreme learning machine model, using variance of its predictions as a correction term accounting for the model uncertainty. The variance is estimated from the model’s linear output layer with a weighted Jackknife method. The methodology is very fast, robust to heteroscedastic outputs, and handles both extremely large datasets and insufficient amount of training data.
|Referentgranskad vetenskaplig tidskrift||International Journal of Machine Learning and Cybernetics|
|Status||Publicerad - 30.01.2018|
|MoE-publikationstyp||A1 Originalartikel i en vetenskaplig tidskrift|
- 512 Företagsekonomi