While recent advances in machine learning (ML) have been immensely successful in several commercial applications with Internet-scale data, the promise of ML is yet to be fully realized for accelerating discoveries in scientific and engineering disciplines. This is because of the rich background of physical knowledge driving real-world phenomena in scientific problems that is currently ignored by mainstream applications of ML. As a result, there is an emerging research trend to deeply integrate scientific knowledge in the ML process, referred to as the paradigm of Science-guided ML (SGML). This course will introduce the foundations of SGML and provide a coherent perspective of research themes in SGML. These research themes will be illustrated using recent examples of cutting-edge research from diverse scientific disciplines. The course will also impart hands-on experience in conducting SGML research through a semester-long project. All course activities will be conducted online.

While there are no formal prerequisites, this course is meant for two categories of graduate students: (a) students familiar in ML who are eager and willing to learn about scientific problems and pursue SGML research, and (b) students from scientific disciplines with little familiarity in ML who are eager to learn and apply SGML in an area they are familiar with.