Special Collection on Uncertainty Quantification for Machine Learning in Engineering (SC062A)
Submit Paper » Background Understanding the data and reaching accurate conclusions are of paramount importance in the present era of Big Data. Machine learning has been widely used in academia and industry to analyze voluminous and intricate datasets to uncover hidden patterns and reach incisive insights. Whilst machine learning approaches have extraordinary potential and are increasingly employed to aid in various complicated tasks, their results are not wholly reliable due to the challenges introduced by data uncertainty (aleatory uncertainty) and model uncertainty (epistemic uncertainty). It is essential to accommodate uncertainties and provide uncertainty estimates to uncover beneficial information for a better decision-making process. To this end, the development and application of novel uncertainty quantification methods in tandem with different machine-learning-enhanced techniques are crucial to yield useful information and amplify the interpretability and reliability of the results. With this in mind, this SC will gather contributions presenting state-of-the-art breakthroughs in uncertainty quantification for machine learning. ...