Learning for Never-before-seen Biomedicine

Explore a thought-provoking colloquium presentation by Stanford's Sheng Wang on "Learning for Never-before-seen Biomedicine." Delve into the computational challenges underlying various biomedical problems, including COVID-19, cancer early identification, and drug side effects. Discover novel machine learning methods developed to tackle two types of never-before-seen situations: never-before-seen class and never-before-seen cohort. Learn how large-scale biomedical ontologies are embedded to classify samples into new classes, leading to discoveries in protein functions, cell types, and rare diseases. Understand the innovative approach of using a multiscale biomedical knowledge graph, constructed from millions of scientific papers and experimental associations, to characterize never-before-seen cohorts. Gain insights into future directions for never-before-seen biomedicine and the importance of cross-disciplinary collaboration in computer science fields such as robotics, security, human-computer interaction, computational design, and ubiquitous computing.

Allen School Colloquium: Sheng Wang (Stanford)