Foreword to this special issue: conformal and probabilistic prediction with applications

This issue of the Annals of Mathematics and Artificial Intelligence is devoted to conformal prediction, a modern machine learning technique that provides predictions and wellcalibrated measures of confidence for individual observations without assuming anything more than that the data are generated independently from the same probability distribution. Most of the papers in this issue have been selected from the Fifth Symposium on Conformal and Probabilistic Prediction with Applications, COPA 2016, held in April 2016 at the Centro de Investigaciones Energéticas, Medioambientales y Tecnológicas (CIEMAT), Madrid, Spain. The first Special Issue of this journal devoted to conformal prediction was published in 2015 (Volume 74, Issues 1–2); most of the papers included in that Special Issue were selected from those presented at the First Symposium in the COPA series, held in 2012 in Greece. This new Special Issue summarises the progress of the method of conformal prediction in theory and applications. Its full title is now slightly longer, including the word “probabilistic” to reflect the importance of the method of Venn prediction (a sister technique to conformal prediction also enjoying an automatic property of being well-calibrated) and other methods of probabilistic prediction. Best papers were selected from those presented at the Fifth Symposium, and their authors were asked to revise, update, and extend their papers in order to be published in the AMAI after a new round of reviews. The papers in this Special Issue can be divided into four groups. • The invited paper (the talk being delivered by Vladimir Vapnik):

[1]  Alexander Gammerman,et al.  Conformal prediction of biological activity of chemical compounds , 2017, Annals of Mathematics and Artificial Intelligence.