Answers are grounded in this paper and cite their sources.
In the above discussions [1], [2], some concerns have been raised related to the suitability of coverage-width-criterion (CWC) for comprehensive evaluation of the quality of prediction intervals (PIs) generated using neural networks (NNs). Effects of existing CWC measure on the quality of PIs reported in [3] and [4] are analyzed in this short note. Alternative methods for optimal construction and comprehensive evaluation of PIs are also discussed.
No takes yet. Share an insight, caveat, or question.
Khosravi et al. (2014) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: