Embedding scale: new thinking of scale in machine learning by May Yuan and Arlo McKee, accepted for publication in a special issue on "Scale and Spatial Analytics", Journal of Geographic Systems.

In this paper, we examine embedded scale in data-driven machine learning research, connect the embedding scale to scale operating in the general theory of geographic representation in GIS and scaffold our arguments with a study of using machine learning to detect archaeological features in drone-collected high-density images. Fig3.jpg in this paper originally was published in:Kvamme, K. L. (2018). Experiments in the Automatic Detection of Archaeological Feat ures in Remotely Sensed Data from Great Plains Villages, USA. Paper presented at the CAA 2016.We thank the author for providing a higher quality image and the supporting GIS data (Ceremonial-house.asc, DEM.asc, and Ditch.asc) for use in this study.