To recapitulate, the main contribution of this article is to propose the use of Gaussian process regression
for constructing mortality tables and improvement factors. While being a relatively new machine learning"
paradigm, the underlying statistical methodology and most crucially the software implementation has
matured signicantly in the past decade. To wit, all of the case studies below have been implemented
straightforwardly using publicly-available, free, well-documented software, and required only basic programming
skills. With a much shorter learning curve and enhanced functionality, GP is well-positioned to
be the 21st century framework for mortality data analysis