article-journal

We survey the literature on AI and knowledge work, focusing on whom the technology benefits. We identify emerging divides in who uses and who benefits from AI’s use in knowledge work.

We adapt graph learning methods to incorporate network structure into discrete choice models.

We investigate the challenge of recommending items in multiple non-overlapping organizations and develop targeted training, a surprising simple and effective transfer-learning approach.

We introduce distance metrics on tumor evolutionary trees to use for benchmmarking inference methods or clustering trees.

An extended journal version of our 2018 BIBM paper. We examine how aspects of sequencing data and patterns of tumor evolution affect phylogeny inference.