Prediction intervals for Deep Neural Networks
This addresses uncertainty quantification in deep learning for practitioners, but it is incremental as it adapts an existing method to a new context.
The paper tackles the problem of constructing prediction intervals for neural network outputs by adapting extremely randomized trees to create ensembles, which reduces variance and improves out-of-sample accuracy. Results show superior performance in coverage probability and mean square prediction error compared to state-of-the-art methods like MC dropout and bootstrap.
The aim of this paper is to propose a suitable method for constructing prediction intervals for the output of neural network models. To do this, we adapt the extremely randomized trees method originally developed for random forests to construct ensembles of neural networks. The extra-randomness introduced in the ensemble reduces the variance of the predictions and yields gains in out-of-sample accuracy. An extensive Monte Carlo simulation exercise shows the good performance of this novel method for constructing prediction intervals in terms of coverage probability and mean square prediction error. This approach is superior to state-of-the-art methods extant in the literature such as the widely used MC dropout and bootstrap procedures. The out-of-sample accuracy of the novel algorithm is further evaluated using experimental settings already adopted in the literature.