A Practical Introduction to Machine Learning for Actuaries

Motivation.Supervised Learning - building predictive models based on past examples-is an important part of Machine Learning and contains a vast and ever increasing array of techniques that can be used by Actuaries alongside more traditional methods. Underlying many Supervised Learning techniques are a small number of important concepts which are also relevant to many areas of actuarial practice. In this paper we use the task of predicting aviation incident cause codes to motivate and practically demonstrate these concepts. These concepts will enable Actuaries to structure analysis pipelines to include both traditional and modern Machine Learning techniques, to correctly compare performance and to have increased confidence that predictive models used are optimal.