How do we learn from data? This course shows that virtually every rigorous learning method involves turning abstract problems into concrete math problems, then solving them. Students understand the theoretical principles underlying machine learning methods from simple linear regression to deep neural networks, preparing them for the math and probability found in upper division courses.
This upper division course provides comprehensive exposure to managing relational data and SQL for data science students. Topics include normalization, query optimization, relational algebra, schema design, and RDBMS internals. Students gain hands-on proficiency in handling relational databases effectively.
Students explore inpatient ICU care by examining severe infection management using the MIMIC-IV dataset, a comprehensive database of deidentified ICU patient data. The project teaches healthcare data nuances, EHR systems, and clinical decision making while developing potential products, reports, or health policies. Students gain understanding of the US healthcare system, ICU operations, and the complexities of data science in critical care environments.
I co-designed this lab course that introduces computing fundamentals through MicroPython on BBC MicroBit boards. Students explore using robots as a metaphor for understanding cognitive processes, with discussions and activities focused on the intersection of robotics and cognitive science. The course emphasizes embodied cognition through hands-on hardware programming.
Students work in groups on a hypothetical data science project that mirrors real world workflows. We cover asking data science questions, data ethics, wrangling and accessing data, exploratory analysis, machine learning, and visualization. The course structure follows how an actual data science project unfolds from start to finish.