Structured missing data: Grand challenges in learning from multi-modal data at scale

Missing data are an unavoidable complication in many machine learning tasks. When data are ‘missing at random’ there exist a range of tools and techniques to deal with the issue. However, as machine learning studies become more ambitious, and seek to learn from ever-larger volumes of heterogeneous data, an increasingly encountered problem arises in which missing values exhibit an association or structure, either explicitly or implicitly. Such ‘structured missingness’ raises a range of challenges that have not yet been systematically addressed, and presents a fundamental hindrance to machine learning at scale. 

During this event you will hear from Dr Robin Mitra, Associate Professor of Statistics at UCL and Dr Sarah McGough, Senior Data Scientist at Roche. They will outline the current literature in missing data and introduce the concept of ‘structured missingness’ using both theory and real-world examples in large healthcare datasets. Inspired by these, they will present some of the biggest challenges and opportunities to advance the field of missing data and spotlight ongoing research led by the Turing-Roche partnership.